AU2012200939A1 - Methods for identifying, diagnosing, and predicting survival of lymphomas - Google Patents

Description

S&F Ref: 755464D1 AUSTRALIA PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT Name and Address Government of the United States of America, as of Applicant: represented by Secretary, Department of Health and Human Services, of Office of Technology Transfer National Institutes of Health 6011 Executive Boulevard, Rockville, Maryland, 20852, United States of America Actual Inventor(s): Bruce Tan Louis Staudt George Wright Dave Sandeep Address for Service: Spruson & Ferguson St Martins Tower Level 35 31 Market Street Sydney NSW 2000 (CCN 3710000177) Invention Title: Methods for identifying, diagnosing, and predicting survival of lymphomas The following statement is a full description of this invention, including the best method of performing it known to me/us: 5845c(6020564_1) 1 METHODS FOR IDENTIFYING, DIAGNOSING, AND PREDICTING SURVIVAL OF LYMPHOMAS RELATED APPLICATIONS 5 The present utility application claims priority to provisional patent application U.S. Serial No. 60/500,377 (Staudt et al.), filed September 3, 2003, the disclosure of which is incorporated by reference herein in its entirety, including but not limited to the electronic data submitted on 21 CD-ROMs accompanying the provisional application. 0 FIELD OF THE INVENTION The present invention relates to the field of diagnosing, identifying, and predicting survival in lymphoproliferative disorders. BACKGROUND OF INVENTION A variety of systems for identifying and classifying lymphomas have been 5 proposed over the last 20 years. In the 1980's, the Working Formulation was introduced as a method of classifying lymphomas based on morphological and clinical characteristics. In the 1990's, the Revised European-American Lymphoma (REAL) system was introduced in an attempt to take into account immunophenotypic and genetic characteristics in classifying lymphomas (Harris 1994). The most recent 20 standard, set forth by the World Health Organization (WHO), attempts to build on these previous systems (Jaffe 2001). The WHO classification of lymphomas is based on several factors, including tumor morphology, immunophenotype, recurrent genetic abnormalities, and clinical features. Table 1, below, contains a list of the B and T cell neoplasms that have been recognized by the WHO classification. Each 25 malignancy is listed according to its WHO classification nomenclature, followed by a WHO classification number.

2 Table 1 B-cell neoplasms Category Name WHO ID # Precursor B-cell neoplasms Precursor B-cell lymphoblastic 9835/3 leukemia Precursor B-cell lymphoblastic 9728/3 lymphoma Mature B-cell neoplasms Chronic lymphocytic leukemia 9823/3 Small lymphocytic lymphoma 9670/3 B-cell prolymphocytic leukemia 9833/3 Lymphoplasmacytic lymphoma 9671/3 Splenic marginal zone 9689/3 lymphoma Hairy cell leukemia 9940/3 Plasma cell myeloma 9732/3 Solitary plasmacytoma of bone 9731/3 Extraosseous plasmacytoma 9734/3 Extranodal marginal zone B-cell 9699/3 lymphoma of mucosa associated lymphoid tissue (MALT lymphoma) Nodal marginal zone B-cell 9699/3 lymphoma Follicular lymphoma (Grade 1, 9690/3 2, 3a, 3b) Mantle cell lymphoma 9673/3 Diffuse large B-cell lymphoma 9680/3 Mediastinal (thymic) large B-cell 9679/3 lymphoma Intravascular large B-cell 9680/3 lymphoma Primary effusion lymphoma 9678/3 Burkitt lymphoma 9687/3 Burkitt leukemia 9826/3 B-cel/ proliferations of uncertain Lymphomatoid granulomatosis 9766/1 malignant potential Post-transplant 9970/1 lymphoproliferative disorder, polymorphic T-cell and NK-cell neoplasms Precursor T-cell and NK-cell Precursor T lymphoblastic 9837/3 neoplasms leukemia Precursor T lymphoblastic 9729/3 lymphoma Blastic NK-cell lymphoma 9727/3 Mature T-cell and NK-cell T-cell prolymphocytic leukemia 9834/3 neoplasms T-cell large granular 9831/3 lymphocytic leukemia Aggressive NK-cell leukemia 9948/3 Adult T-cell leukemia/lymphoma 9827/3 Extranodal NK-/T-cell 9719/3 lymphoma, nasal type 3 Enteropathy-type T-cell 9717/3 lymphoma Hepatosplenic T-cell lymphoma 9716/3 Subcutaneous panniculitis-like 9708/3 T-cei lymphoma Mycosis fungoides 9700/3 Sezary syndrome (9701/3) 9701/3 Primary cutaneous anaplastic 9718/3 large cell lymphoma (C-ALCL) Peripheral T-cell lymphoma, 9702/3 unspecified Angioimmunoblastic T-cell 9705/3 lymphoma Anaplastic large cell lymphoma 9714/3 T-cell proliferation of uncertain Lymphomatoid papulosis 9718/3 malignant potential Hodgkin lymphoma Nodular lymphocyte 9659/3 predominant Hodgkin lymphoma Classical Hodgkin lymphoma 9650/3 Classical Hodgkin lymphoma, 9663/3 nodular sclerosis Classical Hodgkin lymphoma, 9651/3 lymphocyte-rich Classical Hodgkin lymphoma, 9652/3 mixed cellularity Classical Hodgkin lymphoma, 9653/3 _ymphocye depleted Other diagnoses that have not been given WHO diagnostic numbers include HIV associated lymphoma, germinal center B cell-like subtype of diffuse large B cell lymphoma, activated B cell-like subtype of diffuse large B-cell lymphoma, follicular 5 hyperplasia (non-malignant), and infectious mononucleosis (non-malignant). Although the WHO classification has proven useful in patient management and treatment, patients assigned to the same WHO diagnostic category often have noticeably different clinical outcomes. In many cases, these different outcomes appear to be due to molecular differences between tumors that cannot be readily 10 observed by analyzing tumor morphology. More precise methods are needed for identifying and classifying lymphomas based on their molecular characteristics.

4 SUMMARY OF THE INVENTION Accurate identification of lymphoma type or subtype in a. subject suffering from a lymphoproliferative disorder is important for developing an appropriate therapeutic strategy. Previous attempts have been made to identify lymphomas 5 using gene expression data obtained using a microarray. However, there is a need in the art for more accurate and predictive methods of analyzing this gene expression data. In addition, there is a need for more specific and efficient methods of obtaining gene expression data. The present invention discloses a novel microarray for obtaining gene 10 expression data to be used in identifying lymphoma types and predicting survival in a subject. The present invention further discloses a variety of methods for analyzing gene expression data obtained from a lymphoma sample, and specific algorithms for predicting survival and clinical outcome in a subject suffering from a lymphoma. One embodiment of the present invention provides a composition 15 comprising the set of probes listed in Table 2, located at the end of the Detailed Description section. Preferably, this composition comprises a microarray. In another embodiment, the present invention provides a method of generating a survival predictor for a particular lymphoma type. In this method, one 20 or more biopsy samples that have been diagnosed as belonging to a particular lymphoma type are obtained. Gene expression data is obtained for these samples, and genes with expression patterns associated with longer or shorter survival are identified. Hierarchical clustering is performed to group these genes into gene expression signatures, and the expression of all genes within each signature are 5 averaged to obtain a gene expression signature value for each signature. These gene expression signature values are then used to generate a multivariate survival predictor. In another embodiment, the present invention provides a method for 5 predicting survival in a follicular lymphoma (FL) subject. In this method, a biopsy sample is obtained from the subject and gene expression data is obtained from the biopsy sample. The expression level of those genes belonging to an immune response-1 or immune response-2 gene expression signature are averaged to generate gene expression signature values for each signature. A survival predictor 10 score is then calculated using an equation: [2.71*(immune response-2 gene expression signature value)] - [2.36*(immune response-1 gene expression signature value)]. A higher survival predictor score is associated with a less favorable outcome. In one embodiment, the gene expression data used in this method is obtained using a microarray. 15 In another embodiment, the present invention provides another method for predicting survival in a follicular lymphoma (FL) subject. In this method, a biopsy sample is obtained from the subject and gene expression data is obtained from the biopsy sample. The expression level of those genes belonging to a B cell differentiation, T-cell, or macrophage gene expression signature are averaged to 20 generate gene expression signature values for each signature. A survival predictor score is then calculated using an equation: [2.053*(macrophage gene expression signature value)] - [2.344*(T-cell gene expression signature value)] - [0.729*(B-cell gene expression signature value)]. A higher survival predictor score is associated with a less favorable outcome. In one embodiment, the gene expression data used 25 in this method is obtained usina a microarrav.

6 In another embodiment, the present invention provides yet another method for predicting survival in a follicular lymphoma (FL) subject. In this method, a biopsy sample is obtained from the subject and gene expression data is obtained from the biopsy sample. The expression level of those genes belonging to a macrophage, T 5 cell, or B-cell differentiation gene expression signature are averaged to generate gene expression signature values for each signature. A survival predictor score is then calculated using an equation: [1.51*(macrophage gene expression signature value)] - [2.1 1*(T-cell gene expression signature value)] - [0.505*(B-cell differentiation gene expression signature value)]. A higher survival predictor score is 10 associated with a less favorable outcome. In one embodiment, the gene expression data used in this method is obtained using a microarray. In another embodiment, the present invention provides a method for predicting survival in a diffuse large B cell lymphoma (DLBCL) subject. In this method, a biopsy sample is obtained from the subject and gene expression data is 15 obtained from the biopsy sample. The expression level of those genes belonging to an ABC DLBCL high, lymph node, or MHC class Il gene expression signature are averaged to generate gene expression signature values for each signature. A survival predictor score is then calculated using an equation: [0.586*(ABC DLBCL high gene expression signature value)] - [0.468*(lymph node gene expression 20 signature value)] - [0.336*(MHC class Il gene expression signature value)]. A higher survival predictor score is associated with a less favorable outcome. In one embodiment, the gene expression data used in this method is obtained using a microarray. In another embodiment, the present invention provides another method for 25 oredictina survival in a diffuse larae B cell lvmvhoma (DLBCL) subject. In this 7 method, a biopsy sample is obtained from the subject and gene expression data is obtained from the biopsy sample. The expression level of those genes belonging to a lymph node, germinal B cell, proliferation, or MHC class 11 gene expression signature are averaged to generate gene expression signature values for each 5 signature. A survival predictor score is then calculated using an equation: [ 0.4337*(lymph node gene expression signature)] + [0.09*(proliferation gene expression signature)] - (0.4144*(germinal center B-cell gene expression signature)] - [0.2006*(MHC class II gene expression signature)]. A higher survival predictor score is associated with a less favorable outcome. In one embodiment, the gene 10 expression data used in this method is obtained using a microarray. In another embodiment, the present invention provides yet another method for predicting survival in a diffuse large B cell lymphoma (DLBCL) subject. In this method, a biopsy sample is obtained from the subject and gene expression data is obtained from the biopsy sample. The expression level of those genes belonging to 15 a lymph node, germinal B cell, or MHC class Il gene expression signature are averaged to generate gene expression signature values for each signature. A survival predictor score is then calculated using an equation: [-0.32*(lymph node gene expression signature)] - [0.1 76*(germinal B cell gene expression signature)] [0.206*(MHC class Il gene expression signature)]. A higher survival predictor score 20 is associated with a less favorable outcome. In one embodiment, the gene expression data used in this method is obtained using a microarray. In another embodiment, the gene expression data is obtained using RT-PCR. In another embodiment, the present invention provides a method for predicting survival in a mantle cell lymphoma (MCL) subject. In this method, a 25 hionsv samole is obtained from the subiect and aene expression data is obtained 8 from the biopsy sample. The expression level of those genes belonging to a proliferation gene expression signature are averaged to generate a gene expression signature value. A survival predictor score is then calculated using an equation: [1.66*(proliferation gene expression signature value)]. A higher survival predictor 5 score is associated with a less favorable outcome. In one embodiment, the gene expression data used in this method is obtained using a microarray. In another embodiment, the present invention provides a method for determining the probability that a sample X belongs to a first lymphoma type or a second lymphoma type. In this method, a set of genes is identified that is 10 differentially expressed between the two lymphoma types in question, and a set of scale factors representing the difference in expression between the lymphoma types for each of these genes are calculated. A series of linear predictor scores are generated for samples belonging to either of the two lymphoma types based on expression of these genes. Gene expression data is then obtained for sample X, 15 and a linear predictor score is calculated for this sample. The probability that sample X belongs to the first lymphoma type is calculated using an equation that incorporates the linear predictor score of sample X and the mean and variance of the linear predictor scores for the known samples of either lymphoma type. In another embodiment, the present invention provides a method for 20 determining the lymphoma type of a sample X In this method, a set of genes is identified that is differentially expressed between a first lymphoma type and a second lymphoma type, and a set of scale factors representing the difference in expression of each of these genes between the two lymphoma types are calculated. A series of linear predictor scores are generated for samples belonging to either of 9 the two lymphoma types based on expression of these genes. Gene expression data is then obtained for sample X, and a linear predictor score is calculated for this sample. The probability that sample X belongs to the first lymphoma type is calculated using an equation that incorporates the linear predictor score of sample X 5 and the mean and variance of the linear predictor scores for the known samples of either lymphoma type. This entire process is then repeated with various lymphoma types being substituted for the first lymphoma type, the second lymphoma type, or both. In another embodiment, the present invention provides another method for 0 determining the lymphoma type of a sample X. In this method, a series of lymphoma type pairs are created, with each pair consisting of a first lymphoma type and a second lymphoma type. For each type pair, gene expression data is obtained for a set of genes, and a series of scale factors representing the difference in expression of each of these genes between the two lymphoma types are calculated. 5 A subset of z genes with the largest scale factors are identified, and a series of linear predictor scores are generated for samples belonging to either of the two lymphoma types. Linear predictor scores are calculated for anywhere from I to z of these genes. The number of genes from I to z that results in the largest difference in linear predictor scores between the two lymphoma types is selected, and gene 20 expression data for these genes is obtained for sample X. A linear predictor score is generated for sample X, and the probability that the sample belongs to the first lymphoma type is calculated using an equation that incorporates the linear predictor score for sample X and the mean and variance of the linear predictor scores for the known samples of either lymphoma type.

10 In another embodiment, the present invention provides another method for determining the lymphoma type of a sample X. In this method, a series of lymphoma type pairs are created, with each pair consisting of a first lymphoma type and a second lymphoma type. For each type pair, gene expression data is obtained 5 for a set of genes, and a series of scale factors representing the difference in expression of each of these genes between the two lymphoma types are calculated. The set of genes is divided into gene-list categories indicating correlation with a gene expression signature. Within each gene-list category, a subset of z genes with the largest scale factors are identified, and a series of linear predictor scores are 0 generated for samples belonging to either of the two lymphoma types. Linear predictor scores are calculated for anywhere from 1 to z of these genes. The number of genes from 1 to z that results in the largest difference in linear predictor scores between the two lymphoma types is selected, and gene expression data for these genes is obtained for sample X. A linear predictor score is generated for 5 sample X, and the probability q that the sample belongs to the first lymphoma type is calculated using an equation that incorporates the linear predictor score for sample X and the mean and variance of the linear predictor scores for the known samples of either lymphoma type. A high probability q indicates that sample X belongs to the first lymphoma type, a low probability q indicates that sample X belongs to the 20 second lymphoma type, and a middle probability q indicates that sample X belongs to neither lymphoma type. The cut-off point between high, middle, and low probability values is determined by ranking samples of known lymphoma type according to their probability values, then analyzing every possible cut-off point between adjacent samples using the equation: 3.99*[(% of first lymphoma type I1 misidentified as second lymphoma type) + (% of second lymphoma type misidentified as a first lymphoma type)] + [(% of first lymphoma type identified as belonging to neither lymphoma type) + (% of second lymphoma type identified as belonging to neither lymphoma type)]. The final cut-off points are those that minimize the value of this 5 equation. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 : Method for identifying lymphoma type. Flow chart depicts a general method for identifying lymphoma type using gene expression data. Figure 2: Survival signature analysis. Flow chart depicts method for developing a 10 lymphoma survival predictor based on gene expression patterns. Figure 3: FL survival data. Survival data for 191 subjects diagnosed with FL. Median age at diagnosis was 51 years (ranging from 23 to 81 years), and the subjects had a median follow-up of 6.6 years (8.1 years for survivors, with a range of <1 to 28.2 years). Figure 4: Hierarchical clustering of survival associated genes in FL samples. Each 15 column represents a single FL sample, while each row represents a single gene. Relative gene expression is depicted according to the scale at the bottom of the figure. The dendrogram to the left indicates the degree to which the expression pattern of each gene is correlated with that of the other genes. The bars indicate sets of coordinately regulated genes defined as gene expression signatures. Genes comprising the immune response-1 20 and immune response-2 gene expression signature are listed on the right. Figure 5: Kaplan-Meier plot of survival in FL samples based on survival predictor scores. 191 FL samples were divided into quartiles based on their 12 survival predictor scores. The survival predictor scores were calculated using the equation: [(2.71 *immune response-2 gene expression signature value)] - [(2.36 x immune response-1 gene expression signature value)]. Figure 6: Kaplan-Meier plot of survival in FL samples based on IPI score. 96 5 FL samples were divided into three groups based on their IPI scores. Figure 7: Kaplan-Meier plot of survival in FL samples with low or high risk IPI scores based on survival predictor scores. 96 FL samples with low risk (left panel) or intermediate risk (right panel) IPI scores were divided into quartiles based on their survival predictor scores. The survival predictor scores were calculated [0 using the equation: [(2.71*immune response-2 gene expression signature value)) [(2.36 x immune response-1 gene expression signature value)]. Figure 8: Kaplan-Meier plot of survival in FL samples based on survival predictor scores. 191 FL samples were divided into quartiles based on their survival predictor scores. The survival predictor scores were calculated using the 15 equation: [2.053*(macrophage gene expression signature value)] - [2.344*(T-cell gene expression signature value)] - [0.729*(B-cell differentiation gene expression signature value)]. Figure 9: Kaplan-Meier plot of survival in FL samples based on survival predictor scores. 191 FL samples were divided into quartiles based on their 20 survival predictor scores. The survival predictor scores were calculated using the equation: [1.51*(macrophage gene expression signature value)] - [2.1 1*(T-cell gene expression signature value)] - [0.505*(B-cell differentiation gene expression signature value)].

13 Figure 10: Kaplan-Meier plot of survival in DLBCL samples based on survival predictor scores. 231 DLBCL samples were divided into quartiles based on their survival predictor scores. The survival predictor scores were calculated using the equation: [0.586*(ABC DLBCL high gene expression signature value)] 5 [0.468*(lymph node gene expression signature value)] - [(0.336*MHC Class 11 gene expression signature value)]. Figure 11: Kaplan-Meier plot of survival in DLBCL samples based on survival predictor scores. 200 DLBCL samples were divided into quartiles based on their survival predictor scores. The survival predictor scores were calculated using the 0 equation: [-0.4337*(lymph node gene expression signature value)] + [0.09*(proliferation gene expression signature value)} - [0.4144*(germinal center B cell gene expression signature value)] - [0.2006*(MHC class I gene expression signature value)]. Figure 12: Kaplan-Meier plot of survival in DLBCL samples based on survival 5 predictor scores. 200 DLBCL samples were divided into quartiles based on their survival predictor scores. The survival predictor scores were calculated using the equation: [-0.32*(Iymph node gene expression signature value)) - [0.176*(germinal center B-cell gene expression signature value)] - [0.206*(MHC class i gene expression signature value)]. 0 Figure 13: Kaplan-Meier plot of survival in MCL samples based on survival predictor scores. 21 MCL samples were divided into two equivalent groups based on their survival predictor scores. The survival predictor scores were calculated using the equation: 1.66*(proliferation gene expression signature value).

14 Figure 14: Kaplan-Meier plot of survival in MCL samples based on survival predictor scores. 21 MCL samples were divided into two equivalent groups based on their survival predictor scores. The survival predictor scores were calculated using the equation: 1.66*(proliferation gene expression signature value). 5 Figure 15: Predicting lymphoma type using Bayesian analysis. Bayes' rule can be used to determine the probability that an unknown sample belongs to a first lymphoma type rather than a second lymphoma type. A linear predictor score is generated for the sample, and the probability that the sample belongs to the first lymphoma type is determined based on the distribution of linear predictor scores within the first and second 1o lymphoma type. Figure 16: Performance of MCL predictor model. Results of the gene-expression based predictor model for MCL are shown for three models (MCL vs. ABC, MCL vs. GCB, MCL vs. SLL). Performance is shown for both the training set and the validation set. is Figure 17: Gene expression-based identification of DLBCL. Expression levels for 27 genes in a subgroup predictor are shown for 274 DLBCL samples. Expression levels are depicted according to the scale shown at the left. The 14 genes used to predict the DLBCL subgroups in the Affymetrix data set are indicated with asterisks. The probabilities that the DLBCL samples belong to the ABC or GCB subtypes are graphed at 20 the top, and the DLBCL cases are arranged accordingly. Cases belonging to either ABC or GCB with 90% or greater probability are indicated. Figure 18: Performance of DLBCL subtype predictor model. Assignments of DLBCL samples to the ABC or GCB subtypes based on hierarchical clustering vs.

15 the predictor model disclosed herein are compared within the training, validation, and total set of samples. Figure 19: Relationship of gene expression in normal B cell subpopulations to DLBCL subtypes. Relative gene expression in the indicated purified B cell populations 5 is depicted according to the scale in Figure 17. The P value of the difference in expression of these genes between the GCB and ABC DLBCL subtypes is shown, and the subtype with the higher expression is shown is indicated (blue, ABC; orange, GCB). A. DLBCL subtype distinction genes that are more highly expressed in germinal center B cells than at other B cell differentiation stages. B. DLBCL subtype distinction genes that are more io highly expressed in plasma cells than at other B cell differentiation stages. Figure 20: Identification of a PMBL gene expression signature. A. Hierarchical clustering identified a set of 23 PMBL signature genes that were more highly expressed in most lymphomas with a clinical diagnosis of PMBL than in lymphomas assigned to the GCB or ABC subtypes. Each row presents gene expression measurements from a single is Lymphochip microarray feature representing the genes indicated. Each column represents a single lymphoma biopsy sample. Relative gene expression is depicted according to the scale shown. B. Hierarchical clustering of the lymphoma biopsy samples based on expression of the PMBL signature genes identified in (A). A "core" cluster of lymphoma cases was identified that highly expressed the PMBL signature genes. 20 Figure 21 : Development of a gene expression-based molecular diagnosis of PMBL. A. A PMBL predictor was created based on expression of the 46 genes shown. Relative gene expression for each lymphoma biopsy sample is presented 16 according to the scale shown in Figure 20. The probability that each sample is PMBL or DLBCL based on gene expression is shown at the top. B. The PMBL predictor was used to classify 274 lymphoma samples as PMBL or DLBCL. Prediction results are summarized on the right, and the relative gene expression for each case that was classified 5 by the predictor as PMBL is shown on the left. Average expression of each gene in samples classified as DLBCL is also shown. The 20 genes listed are those represented on the Lymphochip that were more highly expressed in PMBL than in DLBCL. Not shown are eight genes from the PMBL predictor that were more highly expressed in DLBCL than in PMBL. 10 Figure 22: Clinical characteristics of PMBL patients. Kaplan-Meier plot of overall survival in PMBL, GCB, and ABC patients after chemotherapy. Figure 23: Optimization of gene number in lymphoma predictor. The optimal number of genes for inclusion in the lymphoma type predictor model is that number which generates a maximum t-statistic when comparing the LPS of two samples from is different lymphoma types. Figure 24: LPS distribution among FL and DLBCL/BL samples. Standard and proliferation LPSs for FL (x) and DLBCL/BL (+) samples. Dotted lines indicate standard deviations from the fitted multivariate normal distributions. Figure 25: Determination of cut-off points for lymphoma classification. The cut-off 20 points between samples classified as DLBCL/BL, FL, or unclassified were optimized to minimize the number of samples classified as the wrong lymphoma type. The optimal lower cut-off point was at cpO.49, while the optimal upper cut-off point was at q=0.8 4

.

17 Figure 26: Division of LPSs among FL and DLBCL/FL samples. Illustration of how the cut-off points described in Figure 25 divided the space between the LPSs of FL (x) and DLBCL/BL (+) samples. Figure 27: Lymphoma classification results. Results of lymphoma classification 5 based on gene expression. 100% of SLL, MCL, and FH samples were classified correctly, and only 3% of DLBCL/BL and FL samples were classified incorrectly. Figure 28: DLBCL classification results. Results of DLBCL subtype classification based on gene expression. None of the ABC samples were classified as the wrong subtype, while only one of the BL samples was classified incorrectly. Of the GCB 0 and PMBL samples, only 5% and 6%, respectively, were classified incorrectly. DETAILED DESCRIPTION The following description of the invention is merely intended to illustrate various embodiments of the invention. As such, the specific modifications discussed are not to be construed as limitations on the scope of the invention. It will be 5 apparent to one skilled in the art that various equivalents, changes, and modifications may be made without departing from the scope of the invention, and it us understood that such equivalent embodiments are to be included herein. Gene expression profiling of a cancer cell or biopsy reflects the molecular phenotype of a cancer at the time of diagnosis. As a consequence, the detailed 20 picture provided by the genomic expression patten provides the basis for a new systematic classification of cancers and more accurate predictors of survival and response to treatment. The present invention discloses methods for identifying, diagnosing, and/or classifying a lymphoma, lymphoid malignancy, or lymphoproliferative disorder based on its gene expression patterns. The present 18 invention also discloses methods for predicting survival in a subject diagnosed with a particular lymphoma type or subtype using gene expression data. The information obtained using these methods will be useful in evaluating the optimal therapeutic approach to be employed with regards to a particular subject. 5 The term "lymphoproliferative disorder" as used herein refers to any tumor of lymphocytes, and may refer to both malignant and benign tumors. The terms "lymphoma" and "lymphoid malignancy" as used herein refer specifically to malignant tumors derived from lymphocytes and lymphoblasts. Examples of lymphomas include, but are not limited to, follicular lymphoma (FL), Burkitt 0 lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, lymphoplasmacytic lymphoma, post transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell 5 lymphoma (GCB), activated B cell-like diffuse large B cell lymphoma (ABC) and primary mediastinal B cell lymphoma (PMBL). The phrase "lymphoma type" (or simply "type") as used herein refers to a diagnostic classification of a lymphoma. The phrase may refer to a broad lymphoma class (e.g., DLBCL, FL, MCL, etc.) or to a subtype or subgroup falling within a broad 20 lymphoma class (e.g., GCB DLBCL, ABC DLBCL). The phrase "gene expression data" as used herein refers to information regarding the relative or absolute level of expression of a gene or set of genes in a cell or group of cells. The level of expression of a gene may be determined based on the level of RNA, such as mRNA, encoded by the gene. Alternatively, the level of 19 thereof encoded by the gene. "Gene expression data" may be acquired for an individual cell, or for a group of cells such as a tumor or biopsy sample. The term "microarray," "array," or "chip" refers to a plurality of nucleic acid probes coupled to the surface of a substrate in different known locations. The 5 substrate is preferably solid. Microarrays have been generally described in the art in, for example, U.S. Patent Nos. 5,143,854 (Pirrung), 5,424,186 (Fodor), 5,445,934 (Fodor), 5,677,195 (Winkler), 5,744,305 (Fodor), 5,800,992 (Fodor), 6,040,193 (Winkler), and Fodor et al. 1991. Light-directed, spatially addressable parallel chemical synthesis. Science, 251:767-777. Each of these references is 10 incorporated by reference herein in their entirety. The term "gene expression signature" or "signature" as used herein refers to a group of coordinately expressed genes. The genes making up this signature may be expressed in a specific cell lineage, stage of differentiation, or during a particular biological response. The genes can reflect biological aspects of the tumors in which [5 they are expressed, such as the cell of origin of the cancer, the nature of the non malignant cells in the biopsy, and the oncogenic mechanisms responsible for the cancer (Shaffer 2001). Examples of gene expression signatures include lymph node (Shaffer 2001), proliferation (Rosenwald 2002), MHC class 1l, ABC DLBCL high, B cell differentiation, T-cell, macrophage, immune response-1, immune response-2, 20 and germinal center B cell. The phrase "survival predictor score" as used herein refers to a score generated by a multivariate model used to predict survival based on gene expression. A subject with a higher survival predictor score is predicted to have poorer survival than a subject with a lower survival predictor score.

20 The term "survival" as used herein may refer to the probability or likelihood of a subject surviving for a particular period of time. Alternatively, it may refer to the likely term of survival for a subject, such as expected mean or median survival time for a subject with a particular gene expression pattern. 5 The phrase "linear predictor score" or "LPS" as used herein refers to a score that denotes the probability that a sample belongs to a particular lymphoma type. An LPS may be calculated using an equation such as: LPS(S)= tS, jeG where Sj is the expression of gene from gene set G in a sample S, and t is a scale 0 factor representing the difference in expression of gene j between a first lymphoma type and a second lymphoma type. Alternatively, a linear predictor score may be generated by other methods including but not limited to linear discriminant analysis (Dudoit 2002), support vector machines (Furey 2000), or shrunken centroids (Tibshirani 2002) 5 The phrase "scale factor' as used herein refers to a factor that defines the relative difference in expression of a particular gene between two samples. An example of a scale factor is a t-score generated by a Student's t-test. The phrase "lymphoma subject," wherein "lymphoma" is a specific lymphoma type (e.g., "follicular lymphoma subject"), may refer to a subject that has been 2O diagnosed with a particular lymphoma by any method known in the art or discussed herein. This phrase may also refer to a subject with a known or suspected predisposition or risk of developing a particular lymphoma type. The pattern of expression of a particular gene is closely connected to the biological role and effect of its gene product. For this reason, the systematic study 2 1 of variations in gene expression provides an alternative approach for linking specific genes with specific diseases and for recognizing heritable gene-variations that are important for immune function. For example, allelic differences in the regulatory region of a gene may influence the expression levels of that gene. An appreciation 5 for such quantitative traits in the immune system may help elucidate the genetics of autoimmune diseases and lymphoproliferative disorders. Genes that encode components of the same multi-subunit protein complex are often coordinately regulated. Coordinate regulation is also observed among genes whose products function in a common differentiation program or in the same 0 physiological response pathway. Recent application of gene expression profiling to the immune system has shown that lymphocyte differentiation and activation are accompanied by parallel changes in expression among hundreds of genes. Gene expression databases may be used to interpret the pathological changes in gene expression that accompany autoimmunity, immune deficiencies, cancers of immune 5 cells and of normal immune responses. Scanning and interpreting large bodies of relative gene expression data is a formidable task. This task is greatly facilitated by algorithms designed to organize the data in a way that highlights systematic features, and by visualization tools that represent the differential expression of each gene as varying intensities and hues of 0 color (Eisen 1998). The development of microarrays, which are capable of generating massive amounts of expression data in a single experiment, has greatly increased the need for faster and more efficient methods of analyzing large-scale expression data sets. In order to effectively utilize microarray gene expression data for the identification and diagnosis of lymphoma and for the prediction of survival in 5 lymphoma patients, new algorithms must be developed to identify important 22 information and convert it to a more manageable format. In addition, the microarrays used to generate this data should be streamlined to incorporate probe sets that are useful for diagnosis and survival prediction. Embodiments of the present invention disclose methods and compositions that address both of these 5 considerations. The mathematical analysis of gene expression data is a rapidly evolving science based on a rich mathematics of pattern recognition developed in other contexts (Kohonen 1997). Mathematical analysis of gene expression generally has three goals. First, it may be used to identify groups of genes that are coordinately 0 regulated within a biological system. Second, it may be used to recognize and interpret similarities between biological samples on the basis of similarities in gene expression patterns. Third, it may be used to recognize and identify those features of a gene expression pattern that are related to distinct biological processes or phenotypes. 5 Mathematical analysis of gene expression data often begins by establishing the expression pattern for each gene on an array across n experimental samples. The expression pattern of each gene can be represented by a point in n-dimensional space, with each coordinate specified by an expression measurement in one of the n samples (Eisen 1998). A clustering algorithm that uses distance metrics can then 0 be applied to locate clusters of genes in this n-dimensional space. These clusters indicate genes with similar patterns of variation in expression over a series of experiments. Clustering methods that have been applied to microarray data in the past include hierarchical clustering (Eisen 1998), self-organizing maps (SOMs) (Tamayo 1999), k-means (Tavazoie 1999), and deterministic annealing (Alon 1999). 5 A variety of different algorithms, each emphasizing distinct ordedv-features of 23 the data, may be required to glean the maximal biological insight from a set of samples (Alizadeh 1998). One such algorithm, hierarchical clustering, begins by determining the gene expression correlation coefficients for each pair of the n genes studied. Genes with similar gene expression correlation coefficients are grouped 5 next to one another in a hierarchical fashion. Generally, genes with similar expression patterns under a particular set of conditions encode protein products that play related roles in the physiological adaptation to those conditions. Novel genes of unknown function that are clustered with a large group of functionally related genes are likely to participate in the same biological process. Likewise, the other clustering 10 methods mentioned herein may also group genes together that encode proteins with related biological function. Gene expression maps may be constructed by organizing the gene expression data from multiple samples using any of the various clustering algorithms outlined herein. The ordered tables of data may then be displayed graphically in a 5 way that allows researchers and clinicians to assimilate both the choreography of gene expression on a broad scale and the fine distinctions in expression of individual genes. In such a gene expression map, genes that are clustered together reflect a particular biological function, and are termed gene expression signatures (Shaffer 20 2001). One general type of gene expression signature includes genes that are characteristically expressed in a particular cell type or at a particular stage of cellular differentiation or activation. Another general type of gene expression signature includes genes that are regulated in their expression by a particular biological process such as proliferation, or by the activity of a particular transcription factor or 25 signaling pathway.

24 The pattern of gene expression in a biological sample provides a distinctive and accessible molecular picture of its functional state and identity (DeRisi 1997; Cho 1998; Chu 1998; Holstege 1998; Spellman 1998). Each cell transduces variation in its environment, internal state, and developmental state into readily 5 measured and recognizable variation in gene expression patterns. Two different samples that have related gene expression patterns are therefore likely to be biologically and functionally similar to one another. Some biological processes are reflected by the expression of genes in a gene expression signature, as described above. The expression of gene expression signatures in a particular sample can 0 provide important biological insights regarding its cellular composition and the function of various intracellular pathways within the cells. The present invention discloses a variety of gene expression signatures related to the clinical outcome of lymphoma patients. While several of these signatures share a name with a previously disclosed signature, each of the gene 5 expression signatures disclosed herein comprises a novel combination of genes. For example, the lymph node signature disclosed herein includes genes encoding extracellular matrix components and genes that are characteristically expressed in macrophage, NK, and T cells (e.g., a-Actinin, collagen type IlIl a 1, connective tissue growth factor, fibronectin, KIAA0233, urokinase plasminogen activator). The !0 proliferation signature includes genes that are characteristically expressed by cells that are rapidly multiplying or proliferating (e.g., c-myc, E21G3, NPM3, BMP6). The MHC class I signature includes genes that interact with lymphocytes in order to allow the recognition of foreign antigens (e.g., HLA-DPa, HLA-DQa, HLA-DRa, HLA DRfi). The immune response-1 signature includes genes encoding T cell markers 5 (e.g., CD7, CD81, ITK, LEFI, STAT4), as well as genes that are highly-expressed 25 in macrophages (e.g., ACTNI, TNFSF13B). The immune response-2 signature includes genes known to be preferentially expressed in macrophages and/or dendritic cells (e.g., TLR5, FCGR1A, SEPT10, LGMN, C3AR1). The germinal center B cell signature includes genes known to be overexpressed at this stage of B 5 cell differentiation (e.g., MME, MEF2C, BCL6, LMO2, PRSPAP2, MBD4, EBF, MYBL1. Databases of gene expression signatures have proven quite useful in elucidating the complex gene expression patterns of various cancers. For example, expression of genes from the germinal center B-cell signature in a lymphoma biopsy .0 suggests that the lymphoma is derived from this stage of B cell differentiation. In the same lymphoma biopsy, the expression of genes from the T cell signature can be used to estimate the degree of infiltration of the tumor by host T cells, while the expression of genes from the proliferation signature can be used to quantitate the tumor cell proliferation rate. In this manner, gene expression signatures provide an 5 "executive summary" of the biological properties of a tumor specimen. Gene expression signatures can also be helpful in interpreting the results of a supervised analysis of gene expression data. Supervised analysis generates a long list of genes with expression patterns that are correlated with survival. Gene expression signatures can be useful in assigning these "predictive" genes to functional 20 categories. In building a multivariate model of survival based on gene expression data, this functional categorization helps to limit the inclusion of multiple genes in the model that measure the same aspect of tumor biology. Gene expression profiles can be used to create multivariate models for predicting survival. The methods for creating these models are called "supervised" 15 because they use clinical data to guide the selection of genes to be used in the prognostic classification. For example, a supervised method might identify genes with expression patterns that correlate with the length of overall survival following chemotherapy. The general method used to create a multivariate model for predicting survival may utilize the following steps: 5 1. Identify genes with expression patterns that are univariately associated with a particular clinical outcome using a Cox proportional hazards model. Generally, a univariate p-value of <0.01 is considered the cut-off for significance. These genes are termed "predictor" genes. 2. Within a set of predictor genes, identify gene expression signatures. 0 3. For each gene expression signature that is significantly associated with survival, average the expression of the component genes within this signature to generate a gene expression signature value. 4. Build a multivariate Cox model of clinical outcome using the gene expression signature values. 5 5. If possible, include additional genes in the model that do not belong to a gene expression signature but which add to the statistical power of the model. This approach has been utilized in the present invention to create novel survival prediction models for FL, DLBCL, and MCL. Each of these models generates a survival predictor score, with a higher score being associated with worse clinical 20 outcome. Each of these models may be used separately to predict survival. Alternatively, these models may be used in conjunction with one or more other models, disclosed herein or in other references, to predict survival. A first FL survival predictor was generated using gene expression data obtained using Affymetrix U133A and U133B microarrays. This predictor 25 incorporated immune response-i and immune response-2 gene expression 27 signatures. Fitting the Cox proportional hazards model to the gene expression signature values obtained from these signatures resulted in the following model: Survival predictor score = [(2.71 *Immune response-2 gene expression signature value)] - [(2.36 x immune response-1 5 gene expression signature value)]. A second FL survival predictor was generated using gene expression data obtained using Affymetrix U133A and U133B microarrays. This predictor incorporated macrophage, T-cell, and B-cell differentiation gene expression signatures. Fitting the Cox proportional hazards model to the gene expression [0 signature values obtained from these signatures resulted in the following model: Survival predictor score = [2.053*(macrophage gene expression signature value)] - [2.344*(T-cell gene expression signature value)] - [0.729*(B-cell differentiation gene expression signature value)]. .5 A third FL survival predictor was generated using gene expression data obtained using the Lymph Dx microarray. This predictor incorporated macrophage, T-cell, and B-cell differentiation gene expression signatures. Fitting the Cox proportional hazards model to the gene expression signature values obtained from these signatures resulted in the following model: 20 Survival predictor score = [1.51*(macrophage gene expression signature value)] - [2.11 *(T-cell gene expression signature value)] - [0.505*(B-cell differentiation gene expression signature value)]. A first DLBCL survival predictor was generated using gene expression data 25 obtained using Affymetrix U133A and U133B microarrays. This predictor 28 incorporated ABC DLBCL high, lymph node, and MHC class If gene expression signatures. Fitting the Cox proportional hazards model to the gene expression signature values obtained from these signatures resulted in the following model: Survival predictor score = [0.586*(ABC DLBCL high gene expression 5 signature value)] - [0.468*(lymph node gene expression signature value)] - 0.336*(MHC class I gene expression signature value)]. A second DLBCL survival predictor was generated using gene expression data obtained using the Lymph Dx microarray. This predictor incorporated lymph 0 node, proliferation, germinal center B-cell, and MHC class Il gene expression signatures. Fitting the Cox proportional hazards model to the gene expression signature values obtained from these signatures resulted in the following model: Survival predictor score = [-0.4337*(lymph node gene expression signature value)] + [0.09*(proliferation gene expression 5 signature value)] - [0.4144*(germinal center B cell gene expression signature value)] [0.2006*(MHC class il gene expression signature value)]. A third DLBCL survival predictor was generated using gene expression data 0 obtained using the Lymph Dx microarray. This predictor incorporated lymph node, germinal center B cell, and MHC class Il gene expression signatures. Fitting the Cox proportional hazards model to the gene expression signature values obtained from these signatures resulted in the following model: Survival predictor score = [-0.32*(lymph node gene expression signature 5 value)] - [0.176*(germinal center B-cell gene- 29 expression signature value)] - [0.206*(MHC class 11 gene expression signature value)]. . An MCL survival predictor was generated using gene expression data obtained using Affymetrix U133A, Affymetrix U133B, and Lymph Dx microarrays. 5 This predictor incorporated a proliferation gene expression signature. Fitting the Cox proportional hazards model to the gene expression signature values obtained from these signatures resulted in the following model: Survival predictor score = [1.66*(proliferation gene expression signature value)]. [0 Gene expression data can also be used to diagnose and identify lymphoma types. In an embodiment of the present invention, a statistical method based on Bayesian analysis was developed to classify lymphoma specimens according to their gene expression profiles. This method does not merely assign a tumor to a particular lymphoma type, but also determines the probability that the tumor belongs 5 to that lymphoma type. Many different methods have been formulated to predict cancer subgroups (Golub 1999; Ramaswamy 2001; Dudoit 2002; Radmacher 2002). These methods assign tumors to one of two subgroups based on expression of a set of differentially expressed genes. However, they do not provide a probability of membership in a subgroup. By contrast, the method disclosed herein used Bayes' 0 rule to estimate this probability, thus allowing one to vary the probability cut-off for assignment of a tumor to a particular subgroup. In tumor types in which unknown additional subgroups may exist, the present method allows samples that do not meet the gene expression criteria of known subgroups to fall into an unclassified group with intermediate probability. A cancer subgroup predictor of the type described 5 herein may be used clinically to provide quantitative diagnostic information for an 30 individual cancer patient. This information can in turn be used to provide a predictor of treatment outcome for a particular cancer patient. For any two lymphoma types A and B, there is a set of genes with significantly higher expression in type A than type B, and a set of genes with significantly lower 5 expression in type A than in type B. By observing the expression of these genes in an unknown sample, it is possible to determine to which of the two types the sample belongs. Evaluating the likelihood that a particular sample belongs to one or the other lymphoma type by Bayesian analysis may be done using the following steps: 1) Identify those genes that are most differentially expressed between the two 0 lymphoma types. This can be done by selecting those genes with the largest t-statistic between the two lymphoma types. The genes in this step may be subdivided into gene expression signatures in certain cases, with genes from each signature analyzed separately. 2) Create a series of linear predictor score (LPS) for samples belonging to 5 either lymphoma type. 3) Evaluate the LPS for each sample in a training set, and estimate the distribution of these scores within each lymphoma type according to a normal distribution. 4) Use Bayes' rule to evaluate the probability that each subsequent sample 0 belongs to one or the other lymphoma type. If only two types of lymphoma are being distinguished, then a single probability score is sufficient to discriminate between the two types. However, if more than two lymphoma types are being distinguished, multiple scores will be needed to highlight specific differences between the types.

3 1 In an embodiment of the present invention, a novel microarray entitled the Lymph Dx microarray was developed for the identification and diagnosis of lymphoma types. The Lymph Dx microarray contains cDNA probes corresponding to approximately 2,653 genes, fewer than the number seen on microarrays that have 5 been used previously for lymphoma diagnosis. The reduced number of probes on the Lymph Dx microarray is the result of eliminating genes that are less useful for the identification of lymphoma types and predicting clinical outcome. This reduction allows for simplified analysis of gene expression data. The genes represented on the Lymph Dx microarray can be divided into four broad categories: 1,101 lymphoma 10 predictor genes identified previously using the Affymetrix U133 microarray, 171 outcome predictor genes, 167 new genes not found on the Affymetrix U 133 microarray, and 1,121 named genes. A list of the probe sets on the Lymph Dx microarray is presented in Table 2, located at the end of the Detailed Description section. [5 In an embodiment of the present invention, gene expression data obtained using the Lymph Dx microarray was used to identify and classify lymphomas using Bayesian analysis. This method was similar to that outlined above, but included additional steps designed to optimize the number of genes used and the cut-off points between lymphoma types. A general overview of this method is presented in 20 Figure 1. Each gene represented on the Lymph Dx microarray was placed into one of three gene-list categories based on its correlation with the lymph node or proliferation gene expression signatures: lymph node, proliferation, or standard. These signatures were identified by clustering of the DLBCL cases using hierarchical clustering and centroid-correlation of 0.35. Standard genes were those 25 with expression patterns that did not correlate highly with expression of the lymph 32 node or proliferation signatures. Lymph Dx gene expression data was first used to identify samples as FL, MCL, SLL, FH, or DLBCL/BL, then to identify DLBCL/BL samples as ABC, GCB, PMBL, or BL. For each stage, a series of pair-wise models was created, with each model containing a different pair of lymphoma types (e.g., FL 5 vs. MCL, SLL vs. FH, etc.). For each pair, the difference in expression of each gene on the microarray was measured, and a t-statistic was generated representing this difference. Genes from each gene-list category were ordered based on their t statistic, and those with the largest t-statistics were used to generate a series of LPSs for samples belonging to either lymphoma type. The number of genes used to 10 generate the LPSs was optimized by repeating the calculation using between five and 100 genes from each gene-list category. The number of genes from each category used in the final LPS calculation was that which gave rise to the largest difference in LPS between the two lymphoma types. Once the number of genes in each gene-list category was optimized, four different LPSs were calculated for each 15 sample. The first included genes from the standard gene-list category only, the second included genes from the proliferation and standard gene-list categories, the third included genes from the lymph node and standard gene-list categories, and the fourth included genes from all three categories. The probability q that a sample X belongs to the first lymphoma type of a pair-wise model can then be calculated using 20 an equation: q O#(LPS(X); A, ,c:) #(LPS(X); A1, 6,) +# (LPS(X); A 2 , 2) LPS(X) is the LPS for sample X, #(x; p, a-) is the normal density function with mean p and standard deviation a-, A, and 6, are the mean and variance of the LPSs 33 for samples belonging to the first lymphoma type, and A^, and6 2 are the mean and variance of the LPSs for samples belonging to the second lymphoma type. Samples with high q values were classified as the first lymphoma type, samples with low q values were classified as the second lymphoma type, and samples with middle 5 range q values were deemed unclassified. To determine the proper cut-off point between high, low, and middle q values, every possible cut-off point between adjacent samples was analyzed by an equation: 3.99 * [(% of type 1 misidentified as type 2) + (% of type 2 misidentified as type 1)] + [(% of type 1 unclassified) + (% of type 2 misidentified)]. ,0 This equation was used to favor the assignment of a sample to an "unclassified" category rather than to an incorrect lymphoma type. The final cut-off points were those which minimized this equation. The coefficient of 3.99 was chosen arbitrarily to allow an additional classification error only if the adjustment resulted in four or more unclassified samples becoming correctly classified. The coefficient can be 5 varied to achieve a different set of trade-offs between the number of unclassified and misidentified samples. To ensure that the accuracy of the model was not a result of overfitting, each model was validated by leave-one-out cross-validation. This entailed removing each sample of known lymphoma type from the data one at a time, and then determining 20 whether the model could predict the missing sample. This process confirmed the accuracy of the prediction method. The classification of a lymphoproliferative disorder in accordance with embodiments of the present invention may be used in combination with any other effective classification feature or set of features. For example, a disorder may be 34 classified by a method of the present invention in conjunction with WHO suggested guidelines, morphological properties, histochemical properties, chromosomal structure, genetic mutation, cellular proliferation rates, immunoreactivity, clinical presentation, and/or response to chemical, biological, or other agents. 5 Embodiments of the present invention may be used in lieu of or in conjunction with other methods for lymphoma diagnosis, such as immunohistochemistry, flow cytometry, FISH for translocations, or viral diagnostics. Accurate determination of lymphoma type in a subject allows for better selection and application of therapeutic methods. Knowledge about the exact 0 lymphoma affecting a subject allows a clinician to select therapies or treatments that are most appropriate and useful for that subject, while avoiding therapies that are nonproductive or even counterproductive. For example, CNS prophylaxis may be useful for treating BL but not DLBCL, CHOP treatment may be useful for treating DLBCL but not blastic MCL (Fisher 1993,; Khouri 1998), and subjects with follicular 5 lymphoma frequently receive treatment while subjects with follicular hyperplasia do not. In each of these situations, the lymphoma types or subtypes in question can be difficult to distinguish using prior art diagnostic methods. The diagnostic and identification methods of the present invention allow for more precise delineation between these lymphomas, which simplifies the decision of whether to pursue a 0 particular therapeutic option. Likewise, the survival prediction methods disclosed in the present invention also allow for better selection of therapeutic options. A subject with a very low survival predictor score (i.e., very good prognosis) may not receive treatment, but may instead be subjected to periodic check-ups and diligent observation. As survival predictor scores increase (i.e., prognosis gets worse), 5 subjects may receive more intensive treatments. Those subjects with the highest 35 survival predictor scores (i.e., very poor prognosis) may receive experimental treatments or treatments with novel agents. Accurate survival prediction using the methods disclosed herein provides an improved tool for selecting treatment options and for predicting the likely clinical outcome of those options. 5 Any effective method of quantifying the expression of at least one gene, gene set, or group of gene sets may be used to acquire gene expression data for use in embodiments of the present invention. For example, gene expression data may be measured or estimated using one or more microarrays. The microarrays may be of any effective type, including but not limited to nucleic acid based or antibody based. 10 Gene expression may also be measured by a variety of other techniques, including but not limited to PCR, quantitative RT-PCR, real-time PCR, RNA amplification, in situ hybridization, immunohistochemistry, immunocytochemistry, FACS, serial analysis of gene expression (SAGE) (Velculescu 1995), Northern blot hybridization, or western blot hybridization. 15 Nucleic acid microarrays generally comprise nucleic acid probes derived from individual genes and placed in an ordered array on a support. This support may be, for example, a glass slide, a nylon membrane, or a silicon wafer. Gene expression patterns in a sample are obtained by hybridizing the microarray with the gene expression product from the sample. This gene expression product may be, for 20 example, total cellular mRNA, rRNA, or cDNA obtained by reverse transcription of total cellular mRNA. The gene expression product from a sample is labeled with a radioactive, fluorescent, or other label to allow for detection. Following hybridization, the microarray is washed, and hybridization of gene expression product to each nucleic acid probe on the microarray is detected and quantified using a detection 25 device such as a phosphorimager or scanning confocal microscope. - 36 There are two broad classes of microarrays: cDNA and oligonucleotide arrays. cDNA arrays consist of hundreds or thousands of cDNA probes immobilized on a solid support. These cDNA probes are usually 100 nucleotides or greater in size. There are two commonly used designs for cDNA arrays. The first is the 5 nitrocellulose filter array, which is generally prepared by robotic spotting of purified DNA fragments or lysates of bacteria containing cDNA clones onto a nitrocellulose filter (Southern 1992; Southern 1994; Gress 1996; Pietu 1996). The other commonly used cDNA arrays is fabricated by robotic spotting of PCR fragments from cDNA clones onto glass microscope slides (Schena 1995; DeRisi 1996; .O Schena 1996; Shalon 1996; DeRisi 1997; Heller 1997; Lashkari 1997). These cDNA microarrays are simultaneously hybridized with two fluorescent cDNA probes, each labeled with a different fluorescent dye (typically Cy3 or Cy5). In this format, the relative mRNA expression in two samples is directly compared for each gene on the microarray. Oligonucleotide arrays differ from cDNA arrays in that the probes are 5 20- to 25-mer oligonucleotides. Oligonucleotide arrays are generally produced by in situ oligonucleotide synthesis in conjunction with photolithographic masking techniques (Pease 1994; Lipshutz 1995; Chee 1996; Lockhart 1996; Wodicka 1997). The solid support for oligonucleotide arrays is typically a glass or silicon surface. 20 Methods and techniques applicable to array synthesis and use have been described in, for example, U.S. Patent Nos. 5,143,854 (Pirrung), 5,242,974 (Holmes), 5,252,743 (Barrett), 5,324,633 (Fodor), 5,384,261 (Winkler), 5,424,186 (Fodor), 5,445,934 (Fodor), 5,451,683 (Barrett), 5,482,867 (Barrett), 5,491,074 (Aldwin), 5,527,681 (Holmes), 5,550,215 (Holmes), 5,571,639 (Hubbell), 5,578,832 !5 (Trulson), 5,593,839 (Hubbell), 5,599,695 (Pease), 5,624,711 (Sundberg.), 5,631,734 37 (Stern), 5,795,716 (Chee), 5,831,070 (Pease), 5,837,832 (Chee), 5,856,101 (Hubbell), 5,858,659 (Sapolsky), 5,936,324 (Montagu), 5,968,740 (Fodor), 5,974,164 (Chee), 5,981,185 (Matson), 5,981,956 (Stern), 6,025,601 (Trulson), 6,033,860 (Lockhart), 6,040,193 (Winkler), 6,090,555 (Fiekowsky), and 6,410,229 (Lockhart), 5 and U.S. Patent Application Publication No. 20030104411 (Fodor). Each of the above patents and applications is incorporated by reference herein in its entirety. Microarrays may generally be produced using a variety of techniques, such as mechanical or light directed synthesis methods that incorporate a combination of photolithographic methods and solid phase synthesis methods. Techniques for the [0 synthesis of microarrays using mechanical synthesis methods are described in, for example, U.S. Patent Nos. 5,384,261 (Winkler) and 6,040,193 (Winkler). Although a planar array surface is preferred, the microarray may be fabricated on a surface of virtually any shape, or even on a multiplicity of surfaces. Microarrays may be nucleic acids on beads, gels, polymeric surfaces, fibers such as fiber optics, glass or any 5 other appropriate substrate. See, for example, U.S. Patent Nos. 5,708,153 (Dower); 5,770,358 (Dower); 5,789,162 (Dower); 5,800,992 (Fodor); and 6,040,193 (Winkler), each of which is incorporated by reference herein in its entirety. Microarrays may be packaged in such a manner as to allow for diagnostic use, or they can be an all-inclusive device. See, for example, U.S. Patent Nos. 20 5,856,174 (Lipshutz) and 5,922,591 (Anderson), both of which are incorporated by reference herein in their entirety. Microarrays directed to a variety of purposes are commercially available from Affymetrix (Affymetrix, Santa Clara, CA). For instance, these microarrays may be used for genotyping and gene expression monitoring for a variety of eukaryotic and Z5 prokaryotic species.

38 The following examples are provided to better illustrate the claimed invention and are not to be interpreted as limiting the scope of the invention. To the extent that specific materials are mentioned, it is merely for purposes of illustration and is not intended to limit the invention. One skilled in the art may develop equivalent 5 means or reactants without the exercise of inventive capacity and without departing from the scope of the invention. It will be understood that many variations can be made in the procedures herein described while still remaining within the bounds of the present invention. It is the intention of the inventors that such variations are included within the scope of the invention. 10 EXAMPLES Example 1: Collection and analysis of gene expression data using Affymetrix U133A and U133B microarrays: 568 cell samples representing various forms of human lymphoid malignancies were obtained by biopsy using known methods described in the literature. The 5 samples were reviewed by a panel of hematopathologists and classified into the following lymphoma types based on current diagnostic criteria: 231 diffuse large B cell lymphomas (DLBCL) 191 follicular lymphomas (FL) 26 Burkitt lymphomas (BL) 20 21 mantle cell lymphoma (MCL) 18 follicular hyperplasias (FH) 17 small cell lymphocytic lymphomas (SLL) 16 mucosa-associated lymphoid tissue lymphomas (MALT) 13 splenic lymphomas (Splenic) 39 10 cyclin-D1 negative lymphomas with MCL morphology (CDInegMCL) 9 multiple myeloma (MultMyeloma) 6 lymphoplasmacytic lymphomas (LPC) 4 post-transplant lymphoproliferative disorders (PTLD) 5 3 lymphoblastic lymphomas (Lymbl) 3 nodal marginal zone lymphomas (NMZ) The 231 DLBCL samples were subdivided into the following lymphoma types based on gene expression (see below): 88 germinal center B cell-like (GCB) [0 78 activated B cell-like (ABC) 33 primary mediastinal B cell lymphoma (PMBL) 32 samples for which the subtype could not be determined (UC_DLBCL) The 16 MALT samples were subdivided into the following four group based on tumor origin: 5 9 from the gastric region (MALT gastric) I from the salivary gland (MALT-salivary) 1 from the lung (MALT-lung) 1 from the tonsil (MALT-tonsil) 4 of unknown origin (MALT unk) 20 Each of the 568 cell samples was given a unique sample ID number consisting of the lymphoma type followed by a unique numerical identifier. For example, "ABC_304" refers to an ABC DLBCL sample numbered 304. Cells were purified and RNA was isolated from the purified cells according to known methods described in the literature.

40 Aliquots of RNA from each sample were applied to Affymetrix U133A and Affymetrix U133B microarrays according to standard Affymetrix protocol. The U133A and U133B microarrays are divided into probe sets, with each probe set consisting of up to 69 oligonucleotide probes 25 nucleotides in length. Each probe 5 set represents a distinct human gene. Information pertaining to these microarrays is available at www.affymetrix.com. Each microarray was scanned using an Affymetrix scanner, which records signal intensity for every probe on the microarray. This information can be transformed into summary signal values for each probe set using a number of different algorithms, including MAS 5.0, D-chip (Li 2001), or 10 Bioconductor's RMA algorithms (Irizarry 2003). The images produced by the scanner were evaluated by Affymetrix MAS 5.0 software. The signal value for each probe on the U133A and U133B microarrays was normalized to a target value of 500, and the base-2 log of the normalized values was used for the following analyses. Log-signal files were statistically analyzed using S+ 15 software and the following S+ subtype predictor script: "superopt.all"< function(data,lab,model,genam=NULL,top=5:50,opt.cuts=T,scale=3.99,highcut=o.99,lowcut=O.5,metho d.cut="mean", 20 include=matrix(T,dim (data)[1 1.1),LWO=T,usecor=F,method.comb= 1,method.genes=1,keeper=c(1,rep( 0,incnum-1))) { 25 patnum-dim(data)[2] #number of samples include as.matrix(include) if(lis.logical(LWO)) 4 1 {runsLWO LWOT else 5 {runs-1 :patnum incnum-dim(include)[21 #number of gene subgroups [0 keep getperms (keeper) 15 modnumdim(mode)[2J #number of pairwise models inmod_getindex(rowSums(model=O)>O) #subtypes that are relevant to the paired models init-Subinit(data~Iab,inmod) initial averages and variances cat("testing",mod num,length (runs), 'An") 20 predict matrix(O.5,patnum,modnum) 25 predict apply(model,2,modsetscript,genam=genam,datrn=data,int=init,Iab=lab,top=top,usecor =usecor,opt.cuts~opt.cuts,scale=scale,highcut=highcut,lowcut=lowcut,method .cut=method.cut,include =include, method. comb~m ethod.com b, method.genes =method.genes, keep=keep) if(LWO) 30 predunlist~lapply(runs,LWOscript,lab=Iab,model=model,data~data, init=init,top=top,usecor--us 42 ecor,opt. cuts =opt.cuts, method.cut=method.cut, include=incj ude,method.comb=method .comb,method .g enes=method. genes, genam =genam, scalescae, highcut=h ighcu t, owcut=Iowcut, keep=keep)) dim (pred)_c(modnurn,Iength (runs)) pred-t(pred) 5 pred2_redict[runs,l predfis .na(pred )L-200 set pred!=-1 00 pred2[setlpredset] pred2[is.na(pred2)I -200 10 pred2[prec12==-200LNA predictlruns]_pred2 predict 15 "getperms"< function (keeper) {incnumtlength(keeper) 20 keep2_matrix(O,2 Aincnumincnum) for(i in I :incnum) { keep2[,il_rep(c(rep(0,2 A(incnum-i)),rep(I ,2 A(incnum-i))),2 A(i- 1) forQj in getindex(keeper== 1)) 25 { keep2_keep2keep2[j]==1 ,drop=FI keep2frowSums(keep2)>O, ,drop=F] 30 43 5 "subinit"< function(train,Iabs,inmod) {genum-dim(train)1~ 0 Irnum_dim(train)[2] labtopmax(iabs) sm sq matrix(O,genum ,Iabtop) nsamprep(O,Iabtop) 5 for(i in inmod) (0 nsamp[iLsum(labs==i) if(nsamp[i]>O) { smf,] rowSums(train[Iabs==i,drop=F]) sq[,fLowSums(train~,Iabs==i,drop=FIA2) list(sm=sm sq=sq,nsamp=nsamp) 44 5 "getsca le"_fun ction (sm, sq, nsamp,m odei, method=1I) if(method==1) #straight average 10 { mdi-model=1l ind2_model==-1 smi-rowSums(sm[,indl ,drop=F]) 5 sql rowSums(sq[,indl,drop=Fl) nsampl-sum(nsamp[indl)) mnl-sml/nsampl vrl sql-(mni A2)*nsampl 20 sm2_rowSums(sm,ind2,drop=F]) sq2-rowSums(sq[, ind2,drop=F]) 25 nsamp2-sum(nsamp[ind2]) mn2_sm2lnsamp2 vr2-sq2-(mn2A2)*nsamp2 30 45 #cat("samps',nsampl ,nsamp2) #cat(Q\n index",getindex(indl ),getindex(ind2),'\n varl ",vrI [1: 1 01,'\n var2Xvr2[1 :10])) nt-nsampl +nsamp2-2 5 vrx-(vrl +vr2)/(nt) vry_(vrx*nt+0.7633)/(nt+2.64) scale_(mnl1 -mn2)/sqrt(vry) 10 if(method==2) #average adjsuted for sample size { indlgetindex(model==1) ind2_getindex(model==-1) t[5 nml_nsamp[indl] mnl-sm[,indl ,drop=F] vrlsq[,ind1 ,drop=F] for(i in 1 :length(indl)) 20 { mnl[,iLmnl[,i]tnml[] vrI [ii (vrl [,i]-(mnl [,iIA2)*nml iJ)(nml [i]-i) nm2_nsamp[ind2j 25 mn2_sm[,ind2,drop=F] vr2-sq[,ind2,drop=F] for(i in 1 :length(ind2)) {mn2[,i]_mn2[,qInm2[il vr2[ I] (vr2[, i]-(mn2[,I]A2)*nm2[i])I(nm2[i]I1) 301 46 dif rowMeans(mn1)-rowMeans(mn2) vri rowMeans(vrl) vr2_rowMeans(vr2) 5 if(length(indl)>1) { vrlvrl+rowVars(mn1) } if(length(ind2)>1) { vr2_vr2+rowVars(mn2) 0} scale=dif/sqrt(vrl *(sum(nm 1-1))+vr2*(sum(nm2-1))) } 5 scale[is.na(scale)LO scale } "truncscale"_function(scale,top,genam,include) { cat("genam",length(genam),"\n") #cat("truncscale",sum(scale! =0),) 5 scalellinclude]_0 gnumlength(scale) if(!is.null(genam)) #re order list to remove duplicates { scordorder(-abs(scale)) reord_(1:gnum)[scord] 0 set duplicated(genam[scord]) 47 set reord[set) scale[set]_O #cat(top, sum (scale! =)) S ord_order(order(-abs(scale))) scale[ord>topl]O if(scai e[ord ==top] ==O) {cat('Only',sum (abs(scale)>o),"genes were used in model") I 10 scale "optim"_function(datl dat2,scale,topset,optmeth) .5 { top settopset[order(topset)] sciord-order(-abs(scale)) datxl _dati datx2_dat.2 20 datl_(datl *scale)[sclordlj dat2_(dat2*scale)[sclordj numi~dim(datl)[2] num2_dim(dlat2)[2] 25 cur-O dif_0 vall rep(O,numl) val2-rep(0,num2) for(i in 1: length (topset)) 30{ curx-topsetil 48 if(curx>dim(datl )f1] {cat("too few available geries",curx,dlim (cdl)[l IAn. break 5 dif set (cur+1 ):curx cur-curx vall-vail +colSums(datl [difset,,drop=F]) vaI2_val2-'colSums(dat2[difset,,drop=FI) inix-mean(vall) t0 mn2x-mean(vai2) vrl x-var(vall) vr2x-var(val2) if(optmeth>1) {pxl-dnorm(vallmnl x,sqrt(vrl x)) [5px2-dnorm (val2,mn2x,sqrt(vr2x)) py2_dnorm (val2, mn 1 x,sqrt(vrl x)) pyldnorm(vall ,mn2x,sqrt(vr2x)) if(optmeth==2) { difx-mean(pxl/(pxl +pyl ))4-mean(px2/(px2+py2)) 20 if(optmeth==3) { difx-sum(pxl /(pxl +pyl ))+sum (px2/(px2+py2)) 25 else {difx-(mnlIx-m n2x)A 2/(vrl x*(numn I-1 )+vr2x*(numn2-l)) if(difx>dif) {dif difx 30 inl-nilx 49 mn2_mn2x vrl-vrlx vr2_vr2x valif-vail 5 val2f-vaI2 opt cur cat('opt=',opt,) 10 #cat('opt=",opt,'mnl =",mnl ,"mn2=",mn2,"vrl ,vrl ,"vr2 "vr2,'\n, "scale=", scae,'\n") ord2_order(sclord) scale[ord2>opt]_O cat("%", mean (scalescale! =01>0),) list(scale~scale,opt~opt,val 1 =vall f,vaI2=val2f) 15 "modsetscript"-function (model, genam ,datrn, datst=datrn, in it,Iab, top, usecor,opt.cuts ,scale ,highcut,Ilowc Autnmethod.cut,include, method.comb,m ethod.genes, keep) 20 { modsetl-is.element(ab,getindex(model== 1)) modset2_is.element(iab,getindex(modei==-l)) cat("modsetl ",getindex(modeI== 1),2modset2",getindex(modeI==-l),"\n") sm-init$sm sqjinit$sq 25 nsamp_init$nsamp modsett-modsetl Imodset2 incnumdim(include)2] patnumndim(datst)[2] if(is.null(patnum)) 30 { patnumi1 50 datst-t(datst) #cat('\n", incn um ,"=incn um\n") scl getscale(sm,sq, nsam p, model, method =method. comb) 5 valt-matrix(O,patnum,incnum) val matrix(O,sum(modsetl ),incnum) vaI2_matrix(O,sum(modset2),incnum) foroj in 1 :incnum) {catoj,) O tpmax(top) sci-truncscaie(scil ,tp,genam,include[,jI) sciset-scl!=0 5 if(Iength(top)>O) optx-optim(datrn[sciset,mod setl ],datrn [sclset, modset2, ,scl[scisetl,top, optmeth =method. genes scljsclset]_optx$scale 20 sclset-scl!=0 cat("opt=",optx$opt,,) val ,jlioptx$val I val2f,j]_optx$va[2 25 else {val [j]_poiSumns (datrn (sclset,modsetl 1*scl [sclset]) va!2[,jlcolSums(datrn[sclset,modset2fsc[sclset) mnlmean(vail) 30 mnn2-mean(vaI2) 5 1 sdl_(stdev(vall)) sd2_(stdev(val2)) valt[,j]_colSums(datst~sclset,I*scl[sclsetl) 51 cat('An") cat("v1 ".dim(valt),"v2",dim(vaI I ),"v3",dim(val2),"\n") optmodel(valt,val 1 ,val2,scale=scale,highcut=highcut,lowcut=lowcut method cut=method .cut,ke ep=keep,usecor=usecor,opt.cuts=opt.cuts) [0 "LWO script"-fu nction(rem ,lab, model, data, init,top,u secor,opt.cuts, method .cut, include, method.comb, m 5 ethod. genes, gen am, scale, highcutlowcu t, keep) { remlab-labiremi predict-rep(-1 O,dim(model)[2]) cat('AnLWO", rem,rem lab,) if(labfrem]l) 20 (modchngNULL) else {modchnggetindex(model[remab,]! =0)) if(Iength(modchng)>0) { initnew-ini 25 initnew$sm[,remlabLinitnew$sm [,remlab]-data[,rem] initnew$sqf,remlab]_initnew$sqf,remlabl-data[reM]A 2 in itnew$nsamp[remlab]_in itnew$nsamp[rem lab]- 1 labnew-lab labnew[rem]0O 52 predictlmodchngliapply(m odel[, modch ngdrop=F],2, modsetscriptgenam genamdatrn=data, datst=datal,rem ,drop=F] initlinitnew,Iab=iabnew,top=top,usecor=usecor opt cuts=opt cuts scale~scale , hihuh g tOCtIWtmehdu mehdcuicueic dmto.comb =method comb, 5 method -genes~method .genes, keep=keep) I predict "optmodeI"_function(vait,vallI vaI2,scale,highcut,Iowcut,method.cut,keep usecor opt cuts) { keepn um-dim (keep)[ 1 vallk-matrix(O,dim(vaII )[II,keepnum) [5 val2k matrix(O,dim(val2)f1 J,keepnum) valtk-matrix(O,dim(valt)[1],keepnum) mnl-colMeans(valI) mn2_colMeans(vaI2) mxmod 0 20 for(i in I :keepnum) { set keep[Q,==l mnl1a-mnl1fkeep[i]==1] 25 mn2a-mn2[keep[ij==1] vi vall[,set,drop=F] v2_va2[,set,drop=F] vt vat[,set,drop=F] if(usecor& min (dim (val2)[l ],dim (va I 1 )(I 1)>(3*sum (keep~i]))) 30 { vrxl-var(vI) 53 vrx2-var(v2) else {vrxldiag(colVars(vl)) 5 vrx2_diag(colVars(v2)) p1lldmvnorm(vl ,mnl a,cov=vrxl) p1 2dmvnorm(vl ,mn2a,cov~vrx2) p2l-dmvnorm(v2,mnl a,cov=vrxl) 10 p22-dmvnorm(v2,mn2a,cov=vrx2) pit dmvnorm(vt,mnl a,cov=vrxl) p2t-dmvnorm(vt,mn2a,dov=vrx2) #cat("lengths",length(pl 1 )jength(pl 2),dim(val I k)[ 1 ],length (p2 1I), length (p22),d im (val2k)1 I'An") 15 vall kfJJ_pll1/(pll1+p12) val2k[i~p21 f(p 2 1 +p22) vaitk[,i]_pl tI(pl t+p2t) 20 if(opt.cuts) { xgetoptcut(rbind(val 1 k,vaf2k),c(rep(1 ,dim(val I k)[1 J),rep( 1 ,dim(vai2k)[1 1)),scale=scale stopl a=1 -lowcut,stop2a=lowcut, stop I b=1 h ighcut, stop2b=h ighcut, method =method.cut) 25 idx-xfl] pout rep(O,dim(vattk)[l1) pout[! is.na(valtk[, idx])&(valtk[,idxj<x[2)J_ -1 pout[!is.na(valtk[, idxl)&(Valtk[jdx]>x[3))]_1 30 else 54 { tstcolMeans(1 -vall k)+colMeans(val2k) idx-order(tst)[1] pout-valtk[,idx] } 5 cat("model choice = ",keep[idxj,"\n") if(opt.cuts) { cat(x,"\n") } if(length(pout)==1) 10 { cat("pout",valtk[,idx],pout,"\n") } pout } 5 20 "getoptcut"< function(data,lab,scale=3.99,scale2=scale,stop 1 a=lnf,stop2a=-Infstopl b= lnfstop2b=lnfmethod="mean") { #data is table of predictor scores lab==1 is assoicated with high values #Iab=-l is associated with low values. scale indicates number misclass= I error 25 numidim(data)[1] if(is.null(num1)) { numl1 dataas.matrix(data)

}

55 num2_dim(data)[2] xl_x2_rep(inf,num2) yl j2rep(lnf,num2) for(i in 1:num2) 5 { ord-order(data[,i]) dat2_data[ord,] lab2-lab[ordj nzl_sum(Iab2==-1) nz2_sum(Iab2==I) 10 if(method=='mean') {scani1 cumsum(-(Iab2==-l )/nzl +scale*(lab2==l )/nz2) scan2_cumsum(-scale*(ab2==-1 )fnzl +QIab2==l )Inz2) else 15 { scani1 cum sum (-(Iab2==- 1)+scale*(lab2==1)) scan2_cumsum(-scale t (lab2==-l )+(Iab2==l)) set-max(getindex(dat2 <stop 1 b)) if(! is.na(set)) 20 { scanl1[1:(set-lyj]Inf scan2[dat2>stop2b]_lnf #uat(scanl) xl [ii min(scanl) 25 idxmin(getindex(scanl ==xl Na)) yl [i] ifelse(idx<num 1 ,(dat2lidx]*scale2+dat2[idx+1 ])/Il +scale2),num I) if(yl [iI>stopl a) { yl[ijstopl a if(method=="mean") 56 { x1 fii_scale*mean(dat2[pb2== 1 I<stopl a>-mean(dat2[lab2== 1]<stopla) else 5 { xi [i]_scale*sum(dat2[lab2==l ]<stopl a)-sum(dat2[Iab2==-1 stopol a) 10 if(yl fi<stopl b) { yl [i1 stopl1b} x2[ijmin(scan2) idxgetindex(scan2==x2[]) 15 y2!jl-ifeise(idx<numl1,(dat2[idx]+scale2*dat2[idx+1 ])/(I +scale2),num 1) if(y2[pzstop2a) I y2FpLstop2a if(method=="mean") 20 ( x2(imean(dat2[lab2==1 ]<stop2a)-scale2*mean(dat2[ab2== 1 j<stop2a) else {x2[i]_sum(dat2pab2==1 ]<stop2a)-scale2*sum(dat2pab2==- I ]<stop2a) 25 ff(y2[i)>stop2b) ( y2[iLstop2b} 30 57 #cat(\kn',xl ,"\n",x2,"\n") 5 x3_xl-'-x2 idx getindex(x3==min(x3)) if(iength(idx)>O) { idx-idxlorder(yl [idx]-y2[idx])Jf1] 10 cutl~yfidx] cut2_y2fidx] if(cut2<cutl) { xcut2 cut2_cuti 15 cuti _x out-c(idx,cutl ,cut2,sum((lab== 1 )[data[, idx] <cutl ]),sum((Iab==1 )[da ta[,idxl <cut2l), sum ((lab== 1 )[data[, idx>cut2j),sum ((lab==- 1)[data[,idxI>cutlI) outl5]_outj5-out[41 20 out[7jout(7]-out6I out 25 'getindex'< function(x) {(1: length (x))[x] 30 58 "rowMax"< function(x, na.rm = T, site = F) { ncol <- dim(x)[2] 5 top <- x[, 1) tops <- rep(1, dim(x)[1]) for(i in 2:ncol) { set <- x[, i] > top if(na.rm) { 10 set[is.na(set)] <- F } top[set] <- x(set, i] tops[set] <- i } 15 if(site){ tp <- data.frame(max = top, site tops) } else { tp <- top 20 } tp } 25 "rowMin"< function(x, na.rm = T, site = F) { ncol <- dim(x)[2] 30 top <- x[, 1] 59 tops <- rep(1, dim(x)[1]) for(i in 2:ncol) { set <- x[, i] < top if(na.rm) { 5 setfis.na(set)] <- F } top[set] <- x[set, i] topsiset] <- i } 10 if(site) { tp <- data.frame(min = top, site = tops) } else tp <- top 15 } tp } 20 ###### Runtime Script begins Here GeneData read.table("GeneData.txt",sep="\t",header-T) 25 GenelD_.read.table("GenelD .txt",sep="\t",header=T,row.names=as.character(1:dim(GeneData)(1])) SamplelD_read.table("SamplelD.txt",sep="\t",header=T) 30 60 incO-rep(T,2745) inc0[c(251 3:2561 .2565:2567)]_F in c2 G enelD[, 51>. 35 5 in,3-GeneD,6]>.35 incl_!inc2&!inc3 genam_Gene)DincO,3J include-data.frame(incl ,inc2,inc3)[inc0,I 10 labsSampielD[,4] colapse matrix(F, 162,58) for(i in c(1:14,51:58)) {colapseli,i]_T 15 1 colapsefc(5,51 ,52,53),40]_T colapse[c(5,51 ,52,53,58),5]T colapsefc(7, 12,14), ISLI colapselc(7,1I2),30]' T 20 colapse[c(1,2,6,9,16,162),16LT modset c(4,5,8,1 1,16) 25 nm_length(modset) nm2_nm*(nm-1)/2 model-matrix(0,1 62,nm2) modnam-matrix(0,nm2,2) 30 tpO- 6 1 for(i in 1:(nm-1)) {set 1 colapse[j modsetf i)] foroj in (i+1):nm) { set2_colapse[,modsetj]j 5 if(sum(setl &set2)==Q) { tptp+1 modei[setl,tpj_1 model[set2,tp]_-1 modriamtp, 1 1_modseti] 0 mnodnamftp,2]_modsetjJ I model-model[, 1 :tpl modnami1_mPodnam[1 :tp,1 modset-c(1 ,2,6,9) nm_Iength(moclset) ?5 nm2_rnm*(nm-l)/2 moclel-matrix(O, 162,nm2) modnam-matrix(O,nm2,2) tpOP for(i in 1 :(nm-1)) WO { setl-colapse,modset[i]] 62 foroj in (i+1):nm) { set2_colapse[,modsejjjj if(sum(setl &set2)==O) { tp~~tp1 5 model[setl tpL1l mode![set2,tp]_-1. modnam[tp, 1J_modset[i] modnamftp,2]_modsetl] 10 model model(,1 :tp] 15 model-data.frame(modeil ,model) modnam-modnam[1 :tp,J modnam-rbind (mod naml 1 modnam) 20 modnam-data.frame(1 :dim(modnam)[1],modnam) datGeneDataincO,) x_superopt.all(dat,iabs ,model,genam~genam ,top=5: 1 O,opt.cuts=T,include=include,LWO=T,Iowcut=O. 25 5,highcut=O.99,usecor--T,keeper--c(1 ,O,O)) 30 res-data.frame(SampleID) 63 inn l-c(16,4,5,8,1 1) Sup~x 5 num-dim(x)[1] seta-is.elernent(modnaml, 3],inn 1 )&is.element(modnam [,2], inn 1) out2_matrix(Q,num, 16) out3_matrix(1,num, 16) 0 for(i in 1: 16) { seti_modnam[seta,2]==i set2_modnam~seta,3j==i cat(sum(setl ),sum(set2),ln") 5 { out2[iJ_out2f,iI+rowSums(Sup[,setal[,setl ,drop=F,na.rm=T) out3f, i]_rowMin(data.frame(out3[, iJ,Sup[, setaf[,set l,drop= Fj)) if(sum(set2)>O) { out2[,i] out2[,iI+rowSums(-Sup[,setal[,set2,drop=FI,na.rm=T) 20 out3[,i]_rowMin(data.frame(out3[,i],-Sup[,seta][,set2 ,drop=FJ)) if(sum(setl +set2)==Q) { out3[,i]_0 res-data.frame(res,rowMax(out2,site=T)) num_dim(res)[2j ~0 res[,num+lLres[,numl 64 res[res[,num-1 I<Iength(inn 1)-i ,num+1 ]OP innl-c(1 .6,9,2) numdim(Sup)[11 5 seta-is.element(modnam[,3], inn 1 )&is.element(modnam [,2], innl1) out2-matrix(0,num, 16) out3-matrix(1 ,num, 16) for(i in 1: 16) (seti -modnam[seta,2==i 10 set2_modnam[seta,3j==i cat(sum (setl ),sum (set2),"\n") if(sum(setl )>0) I ou t2[, iLout2[,17+rowSums (Sup[,seta][,setl ,drop=F], na. rm=T) out3[, i]_rowMin(data.fram e(out3[, il,Sup[, seta][,setl ,drop=F])) 15 if(sum(set2)>O) f out2[,i] out2[,i+rowSums(-Sup,seta1[,set2,drop=F],na.rm=T) out3[, lrowMin(data.frame(out3[,i1,-Sup(,seta(,set,drop=F)) 20 if(sum(setl +set2)==0) I out3[,i1L0 25 res-data.frame(res ,rowMax(out2 ,site=T)) num-dim(res)[21 res[,num+lLres[,num] res[res[,num-1j<length(innl )-1 ,num+ 110 30 res-res[,c(1:3,7,1 0)J 65 for(i in 1:dim(res)[1]) { res[i,6]_lswitch(res[i,4]+1,"Unclassified","ABC","BL",,"FH","FL","GCB",,"MCL","PMBL",,"SLL",,,, ,"Agressive") 5 res[i,7]_switch(resi,5]+1,"Unclassified Aggresive","ABC","BL",,"FH","FL","GCB",,"MCL","PMBL",,"SLL",,,,,"Agressive") } res[,8]_res[,6] 10 res[res,6]=="Agressive",8]_res[res[,6]=="Agressive",7] res_res[,c(1:3,6:8)] rames(res)_c("order,"I Dnumber","Path. Diagnosis","Stage. I.Prediction","Stage. 11. Prediction","Final.Pre diction") 15 write.table(res,file="PredictionResults.txt",sep="\t") Although the log-signal values were analyzed using S+ software and the above algorithm, any effective software/algorithm combination may be used. Example 2: Collection of gene expression data using the novel Lymph Dx 20 microarray: The novel Lymph Dx microarray contains cDNA probes corresponding to approximately 2,734 genes. 174 of these are "housekeeping" genes present for quality control, since they represent genes that are most variably expressed across all lymphoma samples. Other genes represented on the microarray were selected 25 for their utility in identifying particular lymphoma samples and predicting survival in those samples. The genes represented on the Lymph Dx microarray can be divided into four broad categories: 1,101 lymphoma predictor genes identified previously 66 using the Affymetrix U133 microarray, 171 outcome predictor genes identified using the Affymetrix U133 microarray, 167 genes not found on the Affymetrix U133 microarray but represented on the Lymphochip microarray (Alizadeh 1999), and 1,121 named genes. The types of genes making up each of these broad categories 5 are summarized in Table 3, below, while the specific genes represented on the Lymph Dx microarray are listed in Table 2, located at the end of the Detailed Descriptions section. Table 3 Gene type Number of genes Lymphoma predictor genes 1101 Subtype specific 763 Lymph node signature 178 Proliferation signature 160 Outcome predictor genes 171 DLBCL 79 FL 81 MCL 11 New genes not on U133 167 Lymphochip lymphoma predictor genes 84 EBV and HHV8 viral genes 18 BCL-2/cyclin 01/INK4a specialty probes 14 Named genes missing from U133 51 Named genes 1121 Protein kinase 440 Interleukin 35 Interleukin receptor 29 Chemokine 51 Chemokine receptor 29 TNF family 26 TNF receptor family 51 Adhesion 45 Surface marker 264 Oncogene/tumor suppressor 49 Apoptosis 46 Drug target 10 Regulatory 46 10 Cell samples representing various forms of human lymphoid malignancy were obtained by biopsy using known methods described in the literature. These 634 67 biopsy samples were reviewed by a panel of hematopathologists and classified into the following lymphoma types based on current diagnostic criteria: 201 diffuse large B-cell lymphomas (DLBCL) 191 follicular lymphomas (FL) 5 60 Burkitt lymphomas (BL) 21 mantle cell lymphomas (MCL) 30 primary mediastinal B cell lymphoma (PMBL) 18 follicular hyperplasias (FH) 18 small cell lymphocytic lymphomas (SLL) [0 17 mucosa-associated lymphoid tissue lymphomas (MALT), including 9 gastric MALTs (GMALT) 16 chronic lymphocytic leukemias (CLL) 13 splenic lymphomas (SPL) 11 lymphoplasmacytic lymphomas (LPC) .5 11 transformed DLBCL (trDLBCL) (DLBCL that arose from an antecedent FL) 10 cyclin D1 negative lymphomas with MCL morphology (CDIN) 6 peripheral T-cell lymphoma (PTCL) 4 post-transplant lymphoproliferative disorders (PTLD) 4 nodal marginal zone lymphomas (NMZ) 20 3 lymphoblastic lymphomas (LBL) Each of the 634 samples was given a unique sample ID number consisting of the lymphoma type followed by a unique numerical identifier. For example, "BL_2032_52748" refers to a Burkitt lymphoma sample with the numerical identifier 2032_52748. Cells were purified and RNA was isolated from the purified cells 25 according to known methods described in the literature.

68 Aliquots of purified RNA from each sample was applied to the Lymph Dx microarrays according to standard Affymetrix microarray protocol. Each microarray was scanned on an Affymetrix scanner. This scanner produced an image of the microarray, which was then evaluated by Affymetrix MAS 5.0 software. The signal 5 intensity for each probe on the microarray can be transformed into summary signal values for each probe set through a number of different algorithms, including but not limited to MAS 5.0, D-chip (Li 2001), or Bioconductor's RMA algorithms (Irizarry 2003). Example 3: Development of a first FL survival predictor using gene expression 0 data from Affymetrix U133A and U133B microarrays: An analytical method entitled Survival Signature Analysis was developed to create survival prediction models for lymphoma. This method is summarized in Figure 2. The key feature of this method is the identification of gene expression signatures. Survival Signature Analysis begins by identifying genes whose 5 expression pattems are statistically associated with survival. A hierarchical clustering algorithm is then used to identify subsets of these genes with correlated expression patterns across the lymphoma samples. These subsets are operationally defined as "survival-associated signatures." Evaluating a limited number of survival-associated signatures mitigates the multiple comparison 0 problems that are inherent in the use of large-scale gene expression data sets to create statistical models of survival (Ransohoff 2004). FL samples were divided into two equivalent groups: a training set (95 samples) for developing the survival prediction model, and a validation set (96 samples) for evaluating the reproducibility of the model. The overall survival of this 69 cohort is depicted in Figure 3. The median age at diagnosis was 51 years (ranging from 23 to 81 years), and the patients had a median follow-up of 6.6 years (8.1 years for survivors, with a range of <1 to 28.2 years). Gene expression data from Affymetrix U133A and U133B microarrays was obtained for each sample. Within the 5 training set, a Cox proportional hazards model was used to identify "survival predictor" genes, which were genes whose expression levels were. associated with long survival (good prognosis genes) or short survival (poor prognosis genes). A hierarchical clustering algorithm (Eisen 1998) was used to identify gene expression signatures within the good and poor prognosis genes according to their expression 0 pattern across all samples. Ten gene expression signatures were observed within either the good prognosis or poor prognosis gene sets (Figure 4). The expression level of every component gene in each of these ten gene expression signatures was averaged to create a gene expression signature value. To create a multivariate model of survival, different combinations of the ten 5 gene expression signature values were generated and evaluated for their ability to predict survival within the training set. Among models consisting of two signatures, an exceptionally strong statistical synergy was observed between one signature from the good prognosis group and one signature from the poor prognosis group. These signatures were deemed "immune response-I" and "immune response-2," 20 respectively, based on the biological function of certain genes within each signature. The immune response-I gene expression signature included genes encoding T cell markers (e.g., CD7, CD8B1, ITK, LEF1, STAT4) and genes that are highly expressed in macrophages (e.g., ACTNI, TNFSF13B). The immune response-I signature is not merely a surrogate for the number of T cells in the FL biopsy sample 25 because many other standard T cell genes (e.g., CD2, CD4, LAT, TRIM.SH2D1A) 70 were not associated with survival. The immune response-2 gene expression signature included genes known to be preferentially expressed in macrophages and/or dendritic cells (e.g., TLR5, FCGR1A, SEPT10, LGMN, C3AR1). Table 4 lists the genes that were used to generate the gene expression signature values for the 5 immune response-1 and immune response-2 signatures. Table 4 Signature UNIQID Unigene ID Build 167 Gene symbol (http://www.ncbi.nlm. nih.gov/UniGene) Immune response-1 1095985 83883 TMEPAI Immune response-1 1096579 117339 HCST Immune response-1 1097255 380144 Immune response-1 1097307 379754 LOC340061 Immune response-1 1097329 528675 TEAD1 Immune response-i 1097561 19221 C20orf112 Immune response-1 1098152 377588 KIAA1450 Immune response-1 1098405 362807 IL7R Immune response-1 1098548 436639 NFIC Immune response-i 1098893 43577 ATP8B2 Immune response-1 1099053 376041 Immune response-1 1100871 48353 Immune response-I 1101004 2969 SKI Immune response-1 1103303 49605 C9orf52 Immune response-i 1107713 171806 Immune response-i 1115194 270737 TNFSF13B Immune response-I 1119251 433941 SEPWI Immune response-1 1119838 469951 GNAQ Immune response-I 1119924 32309 INPP1 Immune response-I 1120196 173802 TBCI D4 Immune response-i 1120267 256278 TNFRSFIB Immune response-i 1121313 290432 HOXB2 Immune response-1 1121406 NA TNFSF12 Immune response-1 1121720 80642 STAT4 Immune response-1 1122956 113987 LGALS2 Immune response-I 1123038 119000 ACTN1 Immune response-I 1123092 437191 PTRF Immune response-1 1123875 428 FLT3LG Immune response-1 1124760 419149 JAM3 Immune response-i 1128356 415792 CIRL Immune response-1 1128395 7188 SEMA4C Immune response-i 1132104 173802 TBCID4 Immune response-1 1133408 12802 DDEF2 Immune response-1 1134069 405667 CD8B1 Immune response-I 1134751 106185 RALGDS Immune response-1 1134945 81897 KIAA1128 Immune response-1 1135743 299558 TNFRSF25 Immune response-i 1135968 119000 ACTN1 Immune response-1 1136048 299558 TNFRSF25 Immune response-1 1136087 211576 ITK Immune response-1 1137137 195464

-FLNA

7 1 Immune response-1 1137289 36972 D7 Immune response-1 1137534 36972 CD7 Immune response-I 1139339 47099 GALNT1 2 Immune response-1 1139461 14770 BIN2 Immune response-1 1140391 44865 LEF1 Immune response-i 1140524 10784 C6orf37 Immune response-1 1140759 298530 RAB27A Immune response-2 1118755 127826 EPOR Immune response-2 1118966 19196 LOC51619 Immune response-2 1121053 1690 FGFBP1 Immune response-2 1121267 334629 SLN Immune response-2 1121331 8980 TESK2 Immune response-2 1121766 396566 MPP3 Immune response-2 1121852 421391 LECT1 Immune response-2 1122624 126378 ABCG4 Immune response-2 1122679 232770 ALOXE3 Immune response-2 1122770 66578 CRHR2 Immune response-2 1123767 1309 CD1A Immune response-2 1123841 389 ADH7 Immune response-2 1126097 498015 Immune response-2 1126380 159408 Immune response-2 1126628. 254321 CTNNA1 Immune response-2 1126836 414410 NEK1 Immune response-2 1127277 121494 SPAM1 Immune response-2 1127519 NA Immune response-2 1127648 285050 Immune response-2 1128483 444359 SEMA4G Immune response-2 1128818 115830 HS3ST2 Immune response-2 1129012 95497 SLC2A9 Immune response-2 1129582 272236 C21orf77 Immune response-2 1129658 58356 PGLYRP4 Immune response-2 1129705 289368 ADAM19 Immune response-2 1129867 283963 G6PC2 Immune response-2 1130003 432799 Immune response-2 1130388 19196 LOC51619 Immune response-2 1131837 156114 PTPNSI Immune response-2 1133843 6682 SLC7A11 Immune response-2 1133949 502092 PSG9 Immune response-2 1134447 417628 CRHR1 Immune response-2 1135117 512646 PSG6 Immune response-2 1136017 1645 CYP4A1I Immune response-2 1137478 315235 ALDOB Immune response-2 1137745 26776 NTRK3 Immune response-2 1137768 479985 Immune response-2 1138476 351874 HLA-DOA Immune response-2 1138529 407604 CRSP2 Immune response-2 1138601 149473 PRSS7 Immune response-2 1139862 251383 CHST4 Immune response-2 1140189 287369 IL22 Immune response-2 1140389 22116 CDC14B Although the immune response-I and immune response-2 gene expression signatures taken individually were not ideal predictors of survival, the binary model 72 formed by combining the two was more predictive of survival in the training set than any other binary model (p<0.001). Using this binary model as an anchor, other signatures were added to the model using a step up procedure (Drapner 1966). Of the remaining eight signatures, only one signature contributed significantly to the 5 model in the training set (p<0.01), resulting in a three-variable model for survival. This model was associated with survival in a highly statistically significant fashion in both the training (p<0.001) and validation sets (p=0.003). However, only the immune response-1 and immune response-2 gene expression signatures contributed to the predictive power of the model in both the training set and the 0 validation set. The predictive power of each of these signatures is summarized in Table 5. Table 5 Gene expression Contribution of Relative risk of death Effect of increased signature signature to model in among patients In expression on validation set (p- validation set (95% survival value) C.I.) Immune response-1 <0.001 0.15 (0.05-0.46) Favorable immune response-2 <0.001 9.35 (3.02-28.9) Poor Based on this information, the third signature was removed from the model and the 15 two-signature model was used to generate a survival predictor score using the following equation: Survival predictor score = [(2.71*immune response-2 gene expression signature value)] - [(2.36 x immune response-1 gene expression signature value)]. ?0 A higher survival predictor score was associated with worse outcome. The two signature model was associated with survival in a statistically significant fashion in both the training set (p<0.001) and the validation set (p<0.001), which demonstrated 73 that the model was reproducible. For the 187 FL samples with available clinical data, the survival predictor score had a mean of 1.6 and a standard deviation of 0.894, with each unit increase in the predictor score corresponding to a 2.5 fold increase in the relative risk of death. Data for all 191 samples is shown in Table 6. 5 Table 6 Sample Set Length of Status Immune Immune Survival ID # follow-up at response-I response-2 predictor (years) follow-up signature value signature value score FL 1073 Training 7.68 Dead 9.20 8.67 1.77 FL 1074 Training 4.52 Dead 9.10 8.57 1.74 FL 1075 Validation 4.52 Dead 8.97 8.69 2.38 FL 1076 Training 3.22 Dead 9.20 8.55 1.44 FL 1077 Training 7.06 Alive 9.80 8.46 -0.20 FL 1078 Training 4.95 Alive 9.32 8.23 0.30 FL 1080 Training 6.05 Alive 9.45 8.94 1.93 FL 1081 Validation 6.61 Alive 9.00 8.22 1.05 FL 1083 Training 10.01 Alive 9.82 8.72 0.47 FL 1085 Validation 8.84 Alive 9.31 8.58 1.29 FL 1086 Validation 1.98 Dead 9.49 9.09 2.22 FL 1087 Training 8.19 Alive 9.98 9.27 1.57 FL 1088 Validation 5.30 Alive 9.22 8.47 1.20 FL 1089 Training 10.72 Alive 9.42 8.35 0.40 FL 1090 Validation 10.20 Alive 9.27 8.37 0.82 FL 1097 Validation 8.79 Dead 9.87 8.92 0.87 FL 1098 Validation 5.34 Dead 9.33 8.81 1.87 FL 1099 Training 7.65 Alive 9.73 9.04 1.54 FL 1102 Validation 13.20 Dead 9.45 8.89 1.79 FL 1104 Training 8.42 Dead 9.30 8.27 0.48 FL 1106 Validation 7.94 Alive 9.13 9.19 3.36 FL 1107 Training 5.01 Dead 9.41 9.32 3.07 FL 1183 Training 11.56 Dead 9.31 8.53 1.16 FL 1184 Training 6.93 Dead 9.66 8.83 1.13 FL 1185 Validation 7.02 Dead 9.23 9.09 2.86 FL 1186 Training 1.34 Dead 9.01 8.84 2.68 FL 1416 Validation 6.21 Alive 9.50 8.67 1.08 FL 1417 Training 2.40 Dead 8.47 8.39 2.73 FL 1418 Validation 3.59 Alive 8.94 8.42 1.72 FL 1419 Training 3.85 Alive 9.82 8.56 0.03 FL 1422 Training 5.72 Alive 9.46 8.49 0.68 FL 1425 Validation 4.26 Alive 8.93 8.50 1.98 FL 1426 Training 7.32 Alive 9.08 8.26 0.97 FL 1427 Training 5.22 Alive 8.57 8.28 2.22 FL 1428 Validation 5.41 Dead 9.22 8.44 1.10 FL 1432 Training 3.66 Alive 9.22 8.95 2.51 FL 1436 Training 9.08 Dead 9.48 8.63 1.02 FL 1440 Training 7.85 Alive 9.07 8.35 1.22 FL 1445 Training 9.24 Dead 8.67 8.66 3.01 FL 1450 Validation 0.65 Dead 9.83 9.99 3.86 FL 1472 Validation 16.72 Alive 8.85 8.49 2.10 FL 1473 Training 15.07 Alive 9.75 8.50 0.02 FL 1474 Validation 2.75 Dead 9.34 9.10 2.62 74 FL 1476 Validation 4.08 Dead 9.51 8.87 1.60 FL 1477 Training 0.59 Dead 9.64 9.06 1.83 FL 1478 Training 12.47 Dead 9.60 8.87 1.39 FL 1479 Training 2.29 Dead 8.71 9.07 4.01 FL 1480 Training 16.29 Alive 9.40 8.67 1.30 FL 1579 Training 8.22 Dead 8.81 8.44 2.10 FL 1580 Training 19.30 Alive 9.58 8.52 0.49 FL 1581 Training 9.52 Dead 9.08 9.02 3.00 FL 1582 Validation 1.30 Dead 8.40 8.18 2.36 FL 1583 Training 15.26 Dead 9.47 8.79 1.48 FL 1584 Training 15.73 Dead 9.44 8.55 0.89 FL 1585 Validation 0.01 Alive 8.96 8.53 1.96 FL 1586 Validation 3.11 Alive 9.38 8.55 1.03 FL 1588 Training 0.49 Dead 9.52 9.06 2.08 FL 1589 Training 3.15 Alive 9.72 8.74 0.72 FL 1591 Training 11.22 Alive 9.49 8.62 0.97 FL 1594 Validation 11.19 Alive 9.25 8.59 1.47 FL 1595 Training 8.03 Alive 9.75 9.60 3.01 FL 1598 Validation 2.80 Dead 8.81 8.33 1.79 FL 1599 Validation 6.17 Alive 9.48 8.65 1.06 FL 1603 Training 5.17 Dead 9.66 9.75 3.63 FL 1604 Training 3.98 Dead 9.24 8.86 2.20 FL 1606 Validation 4.22 Dead 9.45 9.18 2.57 FL 1607 Validation 8.12 Alive 9.40 8.60 1.13 FL 1608 Validation 9.70 Alive 8.92 8.41 1.72 FL 1610 Validation 2.05 Dead 9.33 9.35 3.32 FL 1611 Validation 10.15 Alive 9.42 8.69 1.31 FL 1616 Training 2.36 Dead 9.38 8.82 1.78 FL 1617 Validation 7.85 Alive 8.96 8.49 1.87 FL 1619 Validation 9.24 Dead 9.43 8.56 0.94 FL 1620 Validation 9.36 Dead 9.14 8.35 1.04 FL 1622 Training 14.01 Alive 9.23 8.53 1.33 FL 1623 Training 9.72 Alive 9.67 8.93 1.38 FL 1624 Validation 3.98 Dead 9.05 8.50 1.70 FL 1625 Validation 11.16 Alive 8.98 8.47 1.75 FL 1626 Validation 6.47 Dead 8.59 8.14 1.76 FL 1628 Validation 0.82 Dead 9.80 8.72 0.51 FL 1637 Validation 18.81 Alive 9.95 9.58 2.48 FL 1638 Validation 4.06 Alive 9.13 8.88 2.51 FL 1639 Training 4.75 Alive 9.53 8.89 1.62 FL 1643 Training 0.77 Dead 9.73 9.06 1.58 FL 1644 Validation 3.84 Alive 9.55 8.68 0.98 FL 1645 Training 3.56 Alive 9.49 8.70 1.18 FL 1646 Training 1.97 Dead 9.25 8.61 1.50 FL 1647 Training 1.22 Dead 9.12 8.89 2.55 FL 1648 Training 11.01 Alive 9.13 8.12 0.46 FL 1652 Training 3.72 Dead 9.50 9.14 2.35 FL 1654 Validation 0.30 Dead 8.74 8.28 1.82 FL 1655 Training 8.45 Alive 9.51 8.85 1.53 FL 1656 Validation 9.36 Alive 9.06 8.58 1.87 FL 1657 Training 10.09 Alive 9.53 8.46 0.44 FL 1660 Training 2.32 Alive 8.81 8.38 1.91 FL 1661 Validation 1.48 Alive 9.86 8.90 0.85 FL 1662 Validation 0.74 Dead 9.57 9.15 2.21 FL 1664 Validation 4.53 Dead 9.34 8.62 1.31 FL 1669 Training 4.40 Dead 8.87 8.58 2.30 75 FL 1670 Training 1.88 Alive 9.64 9.45 2.86 FL 1675 Training 4.57 Alive 9.36 8.46 0.84 FL 1681 Validation 4.23 Alive 9.52 8.63 0.91 FL 1683 Validation 4.03 Dead 9.95 9.10 1.19 FL 1684 Training 2.88 Dead 9.53 8.73 1.18 FL 1716 Validation 9.69 Alive 8.95 8.35 1.50 FL 1717 Validation 2.01 Dead 9.35 8.88 1.98 FL 1718 Training 10.35 Alive 9.23 8.13 0.26 FL 1719 Validation 7.70 Dead 9.13 8.50 1.49 FL 1720 Training 3.91 Dead 8.78 8.88 3.33 FL 1729 Training 8.06 Alive 9.35 8.65 1.39 FL 1732 Validation 0.71 Dead 7.81 8.59 4.86 FL 1761 Validation 10.83 Alive 9.31 8.55 1.22 FL 1764 Training 0.42 Dead 9.25 8.87 2.21 FL 1768 Training 13.04 Alive 9.42 8.47 0.72 FL 1771 Training 9.26 Dead 9.09 8.67 2.06 FL 1772 Validation 13.64 Dead 9.49 8.49 0.61 FL 1788 Training 1.00 Dead 9.09 9.13 3.29 FL 1790 Training 1.42 Alive 9.85 9.40 2.22 FL 1792 Validation 2.01 Dead 9.33 8.72 1.61 FL 1795 Training 0.71 Dead 10.19 9.27 1.08 FL 1797 Validation 7.17 Alive 9.34 8.92 2.14 FL 1799 Training 14.18 Alive 9.32 8.63 1.38 FL 1810 Validation 9.91 Alive 8.66 8.41 2.35 FL 1811 Validation 3.04 Alive 9.38 8.27 0.29 FL 1825 Training 2.98 Alive 9.46 9.07 2.25 FL 1827 Training 3.66 Alive 9.80 8.84 0.83 FL 1828 Validation 11.51 Alive 8.99 8.09 0.72 FL 1829 Validation 4.11 Alive 9.57 8.73 1.08 FL 1830 Validation 5.65 Dead 9.01 8.68 2.25 FL 1833 Training 11.95 Alive 9.74 8.67 0.51 FL 1834 Validation 15.92 Alive 9.22 8.72 1.88 FL 1835 Validation 12.49 Alive 9.26 8.83 2.10 FL 1836 Validation 12.24 Alive 9.55 8.64 0.85 FL 1837 Validation 0.55 Dead 9.47 8.84 1.62 FL 1838 Validation 2.54 Alive 9.90 9.12 1.34 FL 1839 Training 4.48 Alive 8.56 8.32 2.34 FL 1841 Training 0.88 Dead 9.32 9.10 2.66 FL 1842 Validation 4.56 Alive 9.73 8.87 1.07 FL 1844 Validation 13.39 Alive 9.41 8.55 0.98 FL 1845 Training 12.92 Dead 9.89 9.04 1.16 FL 1846 Validation 1.80 Dead 9.79 9.61 2.93 FL 1848 Training 12.52 Alive 9.76 8.81 0.82 FL 1851 Training 4.08 Dead 9.43 9.01 2.18 FL 1853 Validation 12.50 Alive 9.28 8.54 1.25 FL 1854 Validation 13.81 Alive 9.32 8.84 1.98 FL 1855 Validation 9.96 Dead 9.31 8.39 0.75 FL 1857 Validation 8.39 Dead 9.80 9.14 1.65 FL 1861 Validation 3.19 Dead 9.47 8.57 0.88 FL 1862 Validation 7.22 Dead 8.96 8.33 1.44 FL 1863 Validation 10.77 Dead 9.31 8.85 2.00 FL 1864 Training 14.25 Alive 9.98 9.12 1.17 FL_1866 Training 10.72 Dead 9.93 8.94 0.79 FL 1870 Validation 6.41 Dead 10.01 9.22 1.36 FL 1873 Training 7.78 Dead 9.39 8.66 1.30 FL 1874 Validation 3.15 Dead 9.38 8.74 1.53 76 FL 1876 Validation 15.07 Alive 9.59 8.72 0.98 FL 1879 Training 7.13 Dead 9.25 8.62 1.53 FL 1880 Validation 12.84 Dead 8.82 8.35 1.82 FL 1882 Training 8.84 Dead 9.43 8.76 1.49 FL 1884 Validation 11.92 Dead 9.48 9.14 2.41 FL 1885 Validation 15.49 Alive 9.70 8.85 1.11 FL 1887 Training 5.14 Dead 9.47 8.57 0.87 FL 1888 Training 15.08 Alive 9.83 8.97 1.11 FL 1890 Training 3.03 Dead 9.29 9.05 2.60 FL 1894 Training 11.37 Dead 9.01 8.64 2.13 FL 1896 Training 12.03 'Alive 9.80 8.56 0.08 FL 1897 Training 9.63 Alive 9.02 8.33 1.29 FL 1898 Training 5.20 Alive 8.82 8.25 1.54 FL 1900 Validation 7.38 Alive 9.13 8.26 0.85 FL 1903 Validation 28.25 Alive 9.07 8.46 1.54 FL 1904 Validation 7.36 Alive 9.16 8.53 1.50 FL 1905 Validation 3.68 Dead 9.25 8.38 0.87 FL 1906 Training 2.35 Dead 8.04 8.69 4.56 FL 1907 Validation 2.35 Dead 8.11 8.21 3.11 FL 1910 Training 13.84 Alive 9.36 8.72 1.56 FL 1912 Validation 0.73 Dead 9.30 9.21 3.02 FL 1913 Training 2.57 Alive 9.77 8.51 0.01 FL 1916 Validation 11.61 Alive 9.22 8.49 1.24 FL 1918 Validation 9.95 Dead 9.54 8.77 1.26 FL 1919 Training 10.84 Dead 9.51 8.81 1.44 FL 735 Validation 11.05 Dead 8.81 8.23 1.53 FL 738 Validation 10.15 Dead 9.19 8.79 2.13 FL 739 Training 10.80 Dead 9.29 8.77 1.85 FL 878 Validation 3.87 r Dead 8.85 8.54 2.26 FL 879 Training 4.34 Dead 8.95 8.74 2.56 FL 886 Validation 3.29 Alive 9.43 8.72 1.40 FL 888 Validation 1.32 Dead 8.76 8.49 2.34 FL 1627 Training NA NA 9.60 8.51 0.40 FL 1429 Training NA NA 8.69 8.28 1.93 FL 1850 Validation NA NA 9.75 8.83 0.92 FL 1735 Validation NA NA 7.32 8.30 5.24 In order to visualize the predictive power of the model, the FL samples were ranked according to their survival predictor scores and divided into four quartiles. Kaplan-Meier plots of overall survival showed clear differences in survival rate in the 5 validation set (Figure 5). The median survival for each of the four quartiles is set forth in Table 7. Table 7 Quartile Median survival (years) 1 13.6 2 11.1 3 10.8 4 3.9 77 Various clinical variables were found to be significantly associated with survival, including the IP and some of its components and the presence of B symptoms. The gene expression-based model was independent of each of these 5 variables at predicting survival. These clinical variables and the relative risk of death associated with each are summarized in Table 8. Table 8 Clinical Criteria % of % of Univariate (clinical Multivariate (clinical variable patients' patients' variable only) variable + survival relative risk of death predictor score) among patients in relative risk of death validation set among patients in validation set Training Validation RR2 (95% p-value RR2 (95% p-value set set C.I.) C.l.) Age 60 64.5 70.2 1.90 0.044 2.21 (1.48- <0.001 >60 35.5 29.8 (1.02- 3.29) 3.56) Stage I-I 33.3 25 1.31 0.447 2.31 (1.51- <0.001 (ll-IV 66.7 75 (0.65- 3.52) 1 -IV_ _2.64) Extranodal 2 5.4 20.2 1.58 0.163 2.21 (1.48- <0.001 sites (#) <2 94.6 79.8 (0.83- 3.30) LDH Normal 77.1 66.2 1.77 0.065 2.40 (1.57- <0.001 Greater 22.9 33.8 (0.97- 3.67) than 3.24) normal ECOG 2 9.4 12.5 2.05 0.090 2.17 (1.40- <0.001 performance (0.89- 3.35) status <2 90.6 87.5 4.71) Gender Male 42 65 1.62 0.105 2.17 (1.45- <0.001 Female 58 35 (0.90- 3.25) 2.90) B-symptoms Present 17.2 21.3 2.05 0.029 2.10 (1.37- <0.001 Absent 82.8 78.7 (31.08- 3.23) Grade 1 45 43.4 N/A 0.118 2.55 (1.63- <0.001 2 34.8 33.3 2.03 3.99) (1.04 3.96) 3 20.2 23.3 1.39 (0.65 2.98) Int'l. Scores 63.1 47.5 N/A 0.029 2.28 (1.46- <0.001 Prognostic 0-1 3.57) Index 4 78 Scores 33.3 45 2.07 2-3 (1.07 4.00) Scores 3.6 7.5 3.73 4-5 (1.18 _ _ I

-

1 11.18) Due to rounding, percentages may not total 100 2 Relative risk of death (RR) based on 2-fold increase in expression 3 RR for grades 2 and 3 calculated with respect to risk of death for grade 1. The p value is calculated for all grades. 5 4 RR for scores 2-3 and 4-5 calculated with respect to risk of death for scores 0-1. The p-value is calculated for all grades. The samples in the validation set were divided into three groups based on their IPI score, and the relationship between survival and IPI score was visualized by 0 Kaplan-Meier plot (Figure 6). Among validation set samples from the low-risk (IPI 0 1) and intermediate risk (IPI 2-3) IPI groups, the gene-expression-based survival predictor could stratify patients into groups differing by more than 5 years with regards to median survival (Figure 7). The high-risk IPI group (IPI 4-5) comprised less than 5% of the samples, and was omitted from this analysis. These results i demonstrate that the gene expression-based model is not merely acting as a surrogate for clinical variables that are known to predict survival in FL, but rather it identifies distinct biological attributes of the tumors that are associated with survival. Example 4: Development of a second FL survival predictor using gene expression data from Affymetrix U133A and U133B microarrays: 0 191 FL were divided into two equivalent groups: a training set (95 samples) for developing the survival prediction model, and a validation set (96 samples) for evaluating the reproducibility of the model. Gene expression data from Affymetrix U 1 33A and U 1 33B microarrays was obtained for each of the samples. A Cox proportional hazards model was used to identify survival predictor genes whose 5 expression levels were associated with long survival (good prognosis genes) or short 79 survival (poor prognosis genes) in the training set. A positive Cox coefficient indicated increasing mortality with increasing expression of the gene, while a negative Cox coefficient indicated decreasing mortality with increasing expression of the gene. 5 A hierarchical clustering algorithm (Eisen 1998) was used to identify gene expression signatures within the good and poor prognosis genes according to their expression pattern across all samples. Eight clusters of coordinately regulated genes were observed within the good prognosis gene set and six clusters were observed in the poor prognosis gene sets. The expression level of every component 0 gene in each of these gene expression signatures was averaged to create a gene expression signature value. After averaging, only ten of the gene expression signatures were found to be significantly associated with survival in the training set (p<0.01). To create a multivariate model of survival, different combinations of these ten gene expression signature averages were generated and evaluated for their 5 ability to predict survival within the training set. Among models consisting of two signatures, an exceptionally strong statistical synergy was noted between one signature from the good prognosis group and one from the poor prognosis group. These gene expression signatures were termed "T-cell" and "macrophage" based on the biological function of certain genes within each signature. The T-cell gene 0 expression signature included genes that were typically expressed in T-cells, while the macrophage gene expression signature included a number of genes typically expressed in macrophages. Although these two signatures taken individually were not the best predictors of survival, the binary model formed by combining the two was more predictive than any combination of three signatures that did not contain 5 these two signatures. Using these two signatures as an anchor, other signatures 80 were added to the model using a step up procedure (Drapner 1966). Only one of the remaining eight signatures, termed the B-cell differentiation signature, contributed significantly to the model in the training set (p=0.054). The B-cell differentiation signature included a number of genes that appear to be involved in B 5 cell signal transduction. Table 9 lists the genes that were used to generate the gene expression signature values for the T-cell, macrophage, and B-cell differentiation gene expression signatures. Table 9 Signature UNIQID Unigene ID Build 167 Gene symbol (http://www.ncbi.nlm. nih.gov/UniGene) B-cell differentiation 1119350 331141 ALDH2 B-cell differentiation 1130922 459987 ANP32B B-cell differentiation 1130923 459987 ANP32B B-cell differentiation 1099291 130774 C9orf105 B-cell differentiation 1102859 446195 FLJ42418 B-cell differentiation 1120976 245644 GCHFR B-cell differentiation 1098862 303669 MGC26694 B-cell differentiation 1111070 202201 B-cell differentiation 1105935 B-cell differentiation 1139017 274424 NANS B-cell differentiation 1108988 3532 NLK B-cell differentiation 1114726 3532 NLK B-cell differentiation 1097897 266175 PAG B-cell differentiation 1097901 266175 PAG B-cell differentiation 1119813 155342 PRKCID B-cell differentiation 1123298 20191 SIAH2 B-cell differentiation 1101439 63335 TERF2 B-cell differentiation 1120316 63335 TERF2 B-cell differentiation 1096035 105794 UGCGL1 T-cell 1134945 81897 KIAA1128 T-cell 1134069 405667 CD81 T-cell 1137809 405667 C381 T-cell 1119251 433941 SEPW1 T-cell 1096579 117339 HCST T-cell 1101004 2969 SKI T-cell . 1137137 195464 FLNA T-cell 1100871 48353 T-cell 1139461 14770 BlN2 T-cell 1128395 7188 SEMA4C T-cell 1119880 442844 FMOD T-cell 1130676 194431 KIAA0992 T-cell 1130668 194431 KIAA0992 T-cell 1135968 119000 ACTN1 T-cell 1097329 528675 TEAD1 T-cell 1098548 436639 NFIC T-cell 1123038 119000 ACTN1 81 T-cell 1128356 415792 C1RL T-cell 1133408 12802 DDEF2 T-cell 1140524 10784 C6orf37 T-cell 1119838 469951 'GNAQ T-cell 1097255 380144 T-cell 1098152 377588 KIAA1450 T-cell 1115194 270737 TNFSF13B T-cell 1124760 419149 JAM3 T-cell 1120267 256278 TNFRSF1B T-cell 1137289 36972 CD7 T-cell 1137534 36972 CD7 T-cell 1097307 379754 LOC340061 T-cell 1123613 97087 CD3Z T-cell 1121720 80642 STAT4 T-cell 1120196 173802 TBC1D4 T-cell 1136087 211576 ITK T-cell 1132104 173802 TBC1D4 T-cel) 1140391 44865 LEF1 T-cell 1098405 362807 IL7R T-cell 1135743 299558 TNFRSF25 T-cell 1136048 299558 TNFRSF25 T-cell 1123875 428 FLT3LG T-cell 1098893 43577 ATP8B2 T-cell 1097561 19221 C20orf112 T-cell 1122956 113987 LGALS2 T-cell 1121406 TNFSF12 T-cell 1125532 T-cell 1138538 2014 TRD T-cell 1103303 49605 C9orf52 T-cell 1119924 32309 INPP1 Macrophage 1123682 114408 TLR5 Macrophage 1099124 355455 SEPT10 Macrophage 1123401 50130 NDN Macrophage 1134379 150833 C4A Macrophage 1137481 150833 C4A Macrophage 1132220 448805 GPRC5B Macrophage 1119400 181046 DUSP3 Macrophage 1131119 349656 SCARB2 Macrophage 1123566 155935 C3AR1 Macrophage 1138443 77424 FCGR1A Macrophage 1127943 9641 C1QA Macrophage 1119998 8986 CIQB Macrophage 1132433 14732 ME1 Macrophage 1119260 18069 LGMN Macrophage 1098278 166017 MITF The three signatures were used to generate a survival predictor score using the following equation: Survival predictor score = [2.053*(macrophage gene expression signature 82 value)] - [2.344*(T-cell gene expression signature value)] - [0.729*(B-cell differentiation gene expression signature value)]. A higher survival predictor score was associated with worse outcome. According to 5 a likelihood ratio test adjusted for the number of variables included, this model was significant in predicting survival in both the training set (p=1.8 x 10~8) and the validation set (p=2.0 x 10-). For the 187 FL samples with available clinical data, the survival predictor score had a mean of -11.9 and a standard deviation of 0.9418, with each unit increase in the predictor score corresponding to a 2.5 fold increase in 0 the relative risk of death. Data for all 191 samples is shown in Table 10. Table 10 Sample Set B cell T-cell Macrophage Survival ID # differentiation signature signature predictor signature value value score - value FL 1073 Training 9.70 9.14 8.58 -10.89 FL 1074 Training 11.11 9.06 8.52 -11.84 FL 1075 Validation 11.23 8.92 8.75 -11.15 FL 1076 Training 10.02 9.21 8.59 -11.25 FL 1077 Training 9.94 9.77 8.44 -12.82 FL 1078 Training 10.67 9.32 8.21 -12.76 FL 1080 Training 10.62 9.44 8.88 -11.64 FL 1081 Validation 10.38 9.00 8.09 -12.04 FL 1083 Training 10.29 9.77 8.74 -12.47 FL 1085 Validation 9.87 9.24 8.43 -11.55 FL 1086 Validation 10.03 9.50 9.02 -11.06 FL 1087 Training 9.83 9.98 9.37 -11.31 FL 1088 Validation 10.57 9.21 8.29 -12.27 FL 1089 Training 10.30 9.38 8.27 -12.53 FL 1090 Validation 9.74 9.24 8.20 -11.93 FL 1097 Validation 9.57 9.82 8.80 -11.93 FL 1098 Validation 11.08 9.40 8.97 -11.69 FL 1099 Training 10.23 9.70 9.12 -11.46 FL 1102 Validation 9.66 9.46 8.90 -10.93 FL 1104 Training 10.72 9.19 8.20 -12.53 FL 1106 Validation 11.11 9.17 9.57 -9.96 FL 1107 Training 9.70 9.42 9.55 -9.54 FL 1183 Training 9.85 9.25 8.44 -11.54 FL 1184 Training 10.12 9.57 8.86 -11.63 FL 1185 Validation 10.75 9.21 9.13 -10.68 FL 1186 Training 9.76 8.88 8.83 -9.80 FL 1416 Validation 9.94 9.45 8.59 -11.77 FL 1417 Training 10.12 8.53 8.43 -10.08 83 FL 1418 Validation 9.35 8.86 8.27 -10.59 FL 1419 Training 10.20 9.76 8.53 -12.81 FL 1422 Training 10.22 9.48 8.40 -12.43 FL 1425 Validation 9.61 8.89 8.58 ~-10.23 FL 1426 Training 10.80 9.06 8.13 -12.41 FL 1427 Training 10.27 8.56 8.13 -10.87 FL 1428 Validation 10.76 9.25 8.38 -12.32 FL 1432 Training 10.51 9.17 9.04 -10.59 FL 1436 Training 9.69 9.40 8.61 -11.42 FL 1440 Training 9.82 9.04 8.21 -11.50 FL 1445 Training 9.24 8.69 8.62 -9.41 FL 1450 Validation 9.70 9.88 10.37 -8.93 FL 1472 Validation 10.78 8.96 8.51 -11.40 FL 1473 Training 9.99 9.70 8.41 -12.75 FL 1474 Validation 10.21 9.27 9.05 -10.59 FL 1476 Validation 9.82 9.44 8.78 -11.27 FL 1477 Training 9.32 9.61 9.03 -10.78 FL 1478 Training 10.19 9.60 8.81 -11.83 FL 1479 Training 10.69 8.78 9.09 -9.71 FL 1480 Training 10.10 9.42 8.70 -11.57 FL 1579 Training 10.15 8.82 8.24 -11.15 FL 1580 Training 10.31 9.59 8.50 -12.54 FL 1581 Training 9.91 8.96 9.05 -9.66 FL 1582 Validation 9.73 8.31 8.06 -10.03 FL 1583 Training 10.95 9.45 8.86 -11.95 FL 1584 Training 9.98 9.38 8.46 -11.89 FL 1585 Validation 10.53 8.88 8.46 -11.11 FL 1586 Validation 10.00 9.30 8.42 -11.81 FL 1588 Training 9.59 9.41 8.94 -10.68 FL 1589 Training 10.29 9.68 8.73 -12.27 FL 1591 Training 10.44 9.45 8.56 -12.18 FL 1594 Validation 10.01 9.25 8.56 -11.41 FL 1595 Training 9.61 9.75 9.65 -10.07 FL 1598 Validation 11.18 8.80 8.31 -11.71 FL 1599 Validation 10.55 9.48 8.60 -12.24 FL 1603 Training 9.40 9.60 9.77 -9.31 FL 1604 Training 9.92 9.21 8.90 -10.54 FL 1606 Validation 9.87 9.45 9.17 -10.52 FL 1607 Validation 9.76 9.37 8.50 -11.63 FL 1608 Validation 9.92 8.90 8.39 -10.85 FL 1610 Validation 10.02 9.38 9.74 -9.30 FL 1611 Validation 10.18 9.41 8.69 -11.64 FL 1616 Training 9.62 9.33 8.85 -10.71 FL 1617 Validation 9.90 8.95 8.39 -10.98 FL 1619 Validation 9.98 9.37 8.47 -11.85 FL 1620 Validation 9.43 8.95 8.12 -1.1.19 FL 1622 Training 9.84 9.15 8.31 -11.56 FL 1623 Training 9.95 9.61 8.97 -11.37 FL 1624 Validation 10.55 9.06 8.43 -11.61 FL 1625 Validation 10.00 8.89 8.23 -11.22 FL 1626 Validation 11.05 8.62 8.10 -11.62 FL 1628 Validation 10.08 9.81 8.66 -12.57 FL 1637 Validation 9.77 9.95 9.59 -10.76 FL 1638 Validation 10.25 9.20 9.07 -10.41 FL 1639 Training 10.29 9.52 8.99 -11.35 FL 1643 Training 9.80 9.72 9.00 -11.46 84 FL 1644 Validation 9.51 9.46 8.61 -11.43 FL 1645 Training 9.39 9.46 8.70 -11.15 FL 1646 Training 9.90 9.25 8.52 -11.42 FL 1647 Training 9.51 9.12 8.95 -9.92 FL 1648 Training 10.02 9.18 7.86 -12.67 FL 1652 Training 9.62 9.39 9.19 -10.16 FL 1654 Validation 10.32 8.59 8.10 -11.02 FL 1655 Training 10.12 9.53 8.75 -11.74 FL 1656 Validation 10.54 9.08 8.55 -11.42 FL 1657 Training 10.53 9.53 8.55 -12.46 FL 1660 Training 10.24 8.75 8.27 -10.99 FL 1661 Validation 10.08 9.85 9.00 -11.97 FL 1662 Validation 9.85 9.56 9.49 -10.11 FL 1664 Validation 10.16 9.35 8.48 -11.92 FL 1669 Training 9.48 8.76 8.28 -10.45 FL 1670 Training 9.76 9.66 9.66 -9.92 FL 1675 Training 10.57 9.28 8.41 -12.18 FL 1681 Validation 10.48 9.52 8.66 -12.19 FL 1683 Validation 9.88 9.92 9.07 -11.83 FL 1684 Training 9.64 9.53 8.85 -11.20 FL 1716 Validation 9.90 8.91 8.22 -11.23 FL 1717 Validation 9.87 9.34 8.95 -10.71 FL 1718 Training 10.00 9.21 7.98 -12.49 FL 1719 Validation 9.87 9.06 8.42 -11.14 FL 1720 Training 10.70 8.77 8.92 -10.05 FL 1729 Training 10.50 9.23 8.65 -11.53 FL 1732 Validation 9.91 7.68 8.54 -7.69 FL 1761 Validation 9.81 9.22 8.39 -11.54 FL 1764 Training 9.81 9.24 8.77 -10.80 FL 1768 Training 10.12 9.36 8.50 -11.86 FL 1771 Training 9.92 9.12 8.68 -10.79 FL 1772 Validation 9.72 9.42 8.43 -11.87 FL 1788 Training 9.65 9.05 9.12 -9.51 FL 1790 Training 9.58 9.83 9.48 -10.56 FL 1792 Validation 9.79 9.29 8.67 -11.11 FL 1795 Training 9.58 10.18 9.33 -11.69 FL 1797 Validation 9.93 9.26 8.79 -10.90 FL 1799 Training 10.49 9.28 8.64 -11.65 FL 1810 Validation 10.06 8.55 8.21 -10.52 FL 1811 Validation 9.84 9.37 8.08 -12.56 FL 1825 Training 10.49 9.44 9.03 -11.24 FL 1827 Training 10.06 9.76 8.84 -12.08 FL 1828 Validation 10.55 8.93 7.67 -12.87 FL 1829 Validation 9.85 9.58 8.65 -11.87 FL 1830 Validation 10.80 8.99 8.67 -11.15 FL 1833 Training 10.41 9.83 8.82 -12.52 FL 1834 Validation 10.81 9.25 8.63 -11.85 FL 1835 Validation 9.36 9.25 8.91 -10.21 FL 1836 Validation 10.58 9.58 8.61 -12.50 FL 1837 Validation 10.22 9.47 8.76 -11.68 FL 1838 Validation 10.51 9.89 9.19 -11.98 FL 1839 Training 10.79 8.54 8.19 -11.09 FL 1841 Training 10.32 9.31 9.18 -10.48 FL 1842 Validation 10.36 9.69 8.92 -11.95 FL 1844 Validation 10.92 9.43 8.49 -12.65 FL 1845 Training 9.87 9.87 9.06 -11.73 85 FL 1846 Validation 9.66 9.81 9.93 -9.63 FL 1848 Training 9.82 9.74 8.70 -12.14 FL 1851 Training 9.89 9.47 9.03 -10.87 FL 1853 Validation 9.96 9.28 8.54 -11.49 FL 1854 Validation 9.97 9.29 8.73 -11.12 FL 1855 Validation 9.95 9.33 8.42 -11.85 FL 1857 Validation 10.35 9.81 9.28 -11.50 FL 1861 Validation 9.73 9.46 8.43 -11.96 FL 1862 Validation 10.42 8.94 8.22 -11.69 FL 1863 Validation 10.79 9.29 8.82 -11.54 FL 1864 Training 9.67 9.97 9.07 -11.80 FL 1866 Training 10.19 9.88 8.89 -12.33 FL 1870 Validation 9.78 10.07 9.30 -11.63 FL 1873 Training 10.09 9.41 8.77 -11.40 FL 1874 Validation 10.05 9.33 8.69 -11.37 FL 1876 Validation 10.15 9.59 8.67 -12.08 FL 1879 Training 9.73 9.21 8.58 -11.06 FL 1880 Validation 10.02 8.79 8.35 -10.77 FL 1882 Training 9.59 9.44 8.80 -11.05 FL 1884 Validation 9.76 9.51 9.26 -10.38 FL 1885 Validation 10.48 9.66 8.75 -12.32 FL 1887 Training 9.98 9.42 8.47 -11.96 FL 1888 Training 9.73 9.83 8.99 -11.67 FL 1890 Training 10.06 9.33 8.98 -10.76 FL 1894 Training 9.85 8.99 8.75 -10.29 FL 1896 Training 10.21 9.80 8.51 -12.94 FL 1897 Training 10.67 8.99 8.26 -11.90 FL 1898 Training 9.59 8.77 8.21 -10.68 FL 1900 Validation 10.12 9.10 8.10 -12.08 FL 1903 Validation 11.08 8.99 8.39 -11.93 FL 1904 Validation 10.20 9.16 8.30 -11.87 FL 1905 Validation 9.73 9.21 8.22 -11.80 FL 1906 Training 9.95 8.15 8.44 -9.01 FL 1907 Validation 10.12 7.95 7.99 -9.62 FL 1910 Training 11.03 9.38 8.74 -12.10 FL 1912 Validation 9.83 9.38 9.36 -9.95 FL 1913 Training 9.81 9.75 8.43 -12.69 FL 1916 Validation 9.83 9.18 8.40 -11.43 FL 1918 Validation 9.86 9.52 8.79 -11.45 FL 1919 Training 9.87 9.53 8.79 -11.48 FL 735 Validation 10.48 8.73 8.23 -11.20 FL 738 Validation 11.05 9.10 8.75 -11.43 FL 739 Training 9.66 9.25 8.74 -10.78 FL 878 Validation 10.61 8.92 8.65 -10.89 FL 879 Training 9.92 8.94 8.78 -10.14 FL 886 Validation 10.16 9.41 8.63 -11.73 FL 888 Validation 9.35 8.76 8.38 -10.15 FL 1627 Training 9.82 9.48 8.49 -11.94 FL 1429 Training 10.06 8.70 8.14 -11.01 FL 1850 Validation 9.58 9.73 8.70 -11.93 FL 1735 Validation 9.60 7.46 8.42 -7.19 In order to visualize the predictive power of the model, the FL samples were ranked according to their survival predictor scores and divided into four quartiles.

86 Kaplan-Meier plots of overall survival showed clear differences in survival rate in the validation set (Figure 8). The median survival for each of the four quartiles is set forth in Table 11. Table 11 Quartile Median survival (yrs.) 5-year survival 10-year survival 1 NR 94% 79% 2 11.6 82% 62% 3 8.8 69% 39% 4 3.9 38% 22% 5 Example 5: Development of a third FL survival predictor using gene expression data from the Lymph Dx microarray: 191 FL samples were divided into two equivalent groups: a training set for developing the survival prediction model, and a validation set for evaluating the D reproducibility of the model. Gene expression data from the Lymph Dx microarray was obtained for those genes listed in Table 9, above. This gene expression data was used to calculate gene expression signature values for the macrophage, T-cell, and B-cell differentiation gene expression signatures, and these signature values were used to generate a survival predictor score using the following equation: 5 Survival predictor score = [1.51*(macrophage gene expression signature value)] - [2.1 1*(T-cell gene expression signature value)] - [0.505*(B-cell differentiation gene expression signature value)]. A higher survival predictor score was associated with worse outcome. For the 187 0 FL samples with available clinical data, the survival predictor score had a mean of 10.1 and a standard deviation of 0.69, with each unit increase in the predictor score corresponding to a 2.7 fold increase in the relative risk of death. Data for all 191 samples is shown in Table 12.

87 Table 12 Sample Set B cell T-cell Macrophage Survival ID # differentiation signature signature predictor signature value value score value FL 1073 Training 8.26 8.17 7.36 -10.30 FL 1074 Training 9.53 8.12 7.56 -10.53 FL 1075 Validation 9.81 8.00 7.99 -9.77 FL 1076 Training 8.46 8.10 7.62 -9.86 FL 1077 Training 8.45 8.66 7.32 -11.49 FL 1078 Training 9.23 8.32 7.32 -11.18 FL 1080 Training 9.18 8.37 7.86 -10.42 FL 1081 Validation 8.96 8.01 6.94 -10.96 FL 1083 Training 8.72 8.65 7.89 -10.75 FL 1085 Validation 8.34 8.17 7.54 -10.07 FL 1086 Validation 8.50 8.35 7.94 -9.94 FL 1087 Training 8.02 8.88 8.48 -10.00 FL 1088 Validation 9.10 8.15 7.38 -10.65 FL 1089 Training 8.76 8.31 7.35 -10.86 FL 1090 Validation 8.18 8.23 7.43 -10.28 FL 1097 Validation 8.07 8.81 7.90 -10.73 FL 1098 Validation 9.53 8.30 8.09 -10.11 FL 1099 Training 8.44 8.56 8.26 -9.86 FL 1102 Validation 7.92 8.43 7.94 -9.80 FL 1104 Training 9.17 8.07 7.21 -10.78 FL 1106 Validation 9.71 8.15 8.77 -8.85 FL 1107 Training 8.16 8.44 8.60 -8.95 FL 1183 Training 8.49 8.15 7.23 -10.56 FL 1184 Training 8.81 8.49 7.91 -10.43 FL 1185 Validation 9.31 8.19 8.06 -9.80 FL 1186 Training 8.43 7.87 7.83 -9.04 FL 1416 Validation 8.42 8.34 7.63 -10.34 FL 1417 Training 8.65 7.51 7.05 -9.58 FL 1418 Validation 7.96 7.82 7.22 -9.62 FL 1419 Training 8.80 8.71 7.55 -11.43 FL 1422 Training 8.63 8.35 7.39 -10.83 FL 1425 Validation 8.21 7.92 7.62 -9.36 FL 1426 Training 9.39 8.09 7.15 -11.01 FL 1427 Training 8.66 7.51 7.00 -9.65 FL 1428 Validation 9.33 8.18 7.39 -10.81 FL 1432 Training 8.98 8.17 7.93 -9.81 FL 1436 Training 8.04 8.17 7.35 -10.20 FL 1440 Training 8.29 7.82 7.15 -9.89 FL 1445 Training 8.04 7.78 7.63 -8.94 FL 1450 Validation 8.25 8.81 9.52 -8.39 FL 1472 Validation 9.29 7.88 7.33 -10.26 FL 1473 Training 8.49 8.57 7.52 -11.03 FL 1474 Validation 8.59 8.09 8.53 -8.54 FL 1476 Validation 8.25 8.39 7.71 -10.23 FL 1477 Training 7.94 8.57 7.88 -10.21 FL 1478 Training 8.57 8.40 7.88 -10.16 FL 1479 Training 9.15 7.83 7.87 -9.27 FL 1480 Training 8.25 8.38 7.44 -10.63 FL 1579 Training 8.70 7.73 7.43 -9.48 FL 1580 Training 8.86 8.46 7.64 -10.79 FL 1581 Training 8.41 7.89 8.08 -8.69 88 FL 1582 Validation 8.20 7.42 6.99 -9.24 FL 1583 Training 9.34 8.34 7.94 -10.32 FL 1584 Training 8.50 8.33 7.75 -10.17 FL 1585 Validation 9.08 7.96 7.72 -9.72 FL 1586 Validation 8.52 8.25 7.36 -10.61 FL 1588 Training 7.97 8.35 7.73 -9.98 FL 1589 Training 8.85 8.48 7.76 -10.66 FL 1591 Training 8.92 8.36 7.77 -10.42 FL 1594 Validation 8.54 8.22 7.74 -9.96 FL 1595 Training 8.05 8.82 8.68 -9.57 FL 1598 Validation 9.74 7.81 6.97 -10.88 FL 1599 Validation 9.13 8.42 7.69 -10.77 FL 1603 Training 7.97 8.66 8.90 -8.86 FL 1604 Training 8.47 8.14 7.75 -9.75 FL 1606 Validation 8.34 8.32 8.11 -9.51 FL 1607 Validation 8.33 8.30 7.39 -10.57 FL 1608 Validation 8.35 7.88 6.98 -10.31 FL 1610 Validation 8.48 8.35 8.86 -8.52 FL 1611 Validation 8.54 8.33 7.64 -10.37 FL 1616 Training 8.03 8.39 7.67 -10.18 FL 1617 Validation 8.30 7.85 7.52 -9.40 FL 1619 Validation 8.53 8.31 7.64 -10.32 FL 1620 Validation 8.09 7.99 7.17 -10.11 FL 1622 Training 8.14 8.10 7.36 -10.09 FL 1623 Training 8.45 8.52 8.15 -9.93 FL 1624 Validation 9.13 8.12 7.46 -10.49 FL 1625 Validation 8.53 7.94 7.17 -10.23 FL 1626 Validation 9.63 7.67 7.17 -10.22 FL 1628 Validation 8.63 8.76 7.95 -10.86 FL 1637 Validation 8.07 8.81 8.79 -9.38 FL 1638 Validation 8.52 8.18 8.19 -9.18 FL 1639 Training 8.70 8.33 7.89 -10.06 FL 1643 Training 8.26 8.62 8.01 -10.26 FL 1644 Validation 8.28 8.33 7.77 -10.02 FL 1645 Training. 7.84 8.32 7.68 -9.91 FL 1646 Training 8.40 8.26 7.71 -10.01 FL 1647 Training 8.10 8.04 7.92 -9.10 FL 1648 Training 8.33 8.08 6.87 -10.90 FL 1652 Training 8.15 8.33 8.37 -9.07 FL 1654 Validation 8.67 7.62 7.03 -9.85 FL 1655 Training 8.53 8.41 7.75 -10.36 FL 1656 Validation 9.09 8.09 7.62 -10.16 FL 1657 Training 8.95 8.44 7.58 -10.89 FL 1660 Training 8.82 7.79 7.26 -9.93 FL 1661 Validation 8.56 8.79 8.17 -10.53 FL 1662 Validation 8.30 8.47 8.69 -8.93 FL 1664 Validation 8.62 8.23 7.56 -10.31 FL 1669 Training 7.89 7.67 7.39 -9.02 FL 1670 Training 8.01 8.54 8.64 -9.03 FL 1675 Training 9.00 8.21 7.36 -10.76 FL 1681 Validation 8.83 8.39 7.59 -10.72 FL 1683 Validation 8.14 8.85 7.97 -10.74 FL 1684 Training 7.99 8.42 7.84 -9.97 FL 1716 Validation 8.28 7.90 7.26 -9.88 FL 1717 Validation 8.27 8.21 7.89 -9.60 FL 1718 Training 8.50 8.17 7.15 -10.75 89 FL 1719 Validation 8.35 8.02 7.21 -10.26 FL 1720 Training 9.03 7.65 8.01 -8.61 FL 1729 Training 8.97 8.27 7.69 -10.37 FL 1732 Validation 8.49 6.82 7.71 -7.02 FL 1761 Validation 8.36 8.19 7.29 -10.49 FL 1764 Training 8.52 8.24 7.94 -9.69 FL 1768 Training 8.70 8.25 7.63 -10.28 FL 1771 Training 8.55 8.19 7.65 -10.04 FL 1772 Validation 8.30 8.38 7.41 -10.71 FL 1788 Training 8.14 8.06 8.11 -8.87 FL 1790 Training 7.95 8.69 8.36 -9.74 FL 1792 Validation 8.16 8.20 7.64 -9.88 FL 1795 Training 7.94 9.08 8.37 -10.54 FL 1797 Validation 8.17 8.21 7.87 -9.57 FL 1799 Training 9.02 8.21 7.77 -10.14 FL 1810 Validation 8.43 7.52 7.06 -9.47 FL 1811 Validation 8.33 8.24 7.07 -10.93 FL 1825 Training 8.90 8.39 7.97 -10.18 FL 1827 Training 8.47 8.77 7.96 -10.76 FL 1828 Validation 9.13 7.87 6.76 -11.01 FL 1829 Validation 8.34 8.51 7.59 -10.71 FL 1830 Validation 9.26 8.04 7.62 -10.13 FL 1833 Training 8.82 8.86 7.88 -11.26 FL 1834 Validation 9.25 8.17 7.62 -10.39 FL 1835 Validation 7.71 8.16 8.01 -9.02 FL 1836 Validation 9.06 8.52 7.59 -11.09 FL 1837 Validation 8.57 8.33 7.37 -10.79 FL 1838 Validation 8.78 8.72 8.04 -10.69 FL 1839 Training 9.27 7.36 7.37 -9.08 FL 1841 Training 8.66 8.35 8.17 -9.64 FL 1842 Validation 8.62 8.50 8.02 -10.19 FL 1844 Validation 9.37 8.40 7.47 -11.18 FL 1845 Training 8.33 8.84 8.30 -10.32 FL 1846 Validation 8.11 8.75 9.06 -8.89 FL 1848 Training 8.19 8.60 7.91 -10.33 FL 1851 Training 8.37 8.50 8.15 -9.84 FL 1853 Validation 8.37 8.14 7.43 -10.19 FL 1854 Validation 8.50 8.29 7.96 -9.78 FL 1855 Validation 8.63 8.34 7.54 -10.58 FL 1857 Validation 8.73 8.82 8.45 -10.26 FL 1861 Validation 8.21 8.50 7.50 -10.77 FL 1862 Validation 8.98 7.96 7.31 -10.28 FL 1863 Validation 9.30 8.22 7.86 -10.18 FL 1864 Training 8.13 8.93 8.27 -10.46 FL 1866 Training 8.62 8.78 7.91 -10.93 FL 1870 Validation 8.16 8.97 8.52 -10.18 FL 1873 Training 8.55 8.30 8.00 -9.74 FL 1874 Validation 8.43 8.20 7.59 -10.10 FL 1876 Validation 8.48 8.52 7.70 -10.64 FL 1879 Training 8.29 8.21 7.66 -9.94 FL 1880 Validation 8.56 7.76 7.34 -9.61 FL 1882 Training 8.02 8.40 7.71 -10.14 FL 1884 Validation 8.14 8.46 8.42 -9.24 FL 1885 Validation 8.88 8.57 7.78 -10.81 FL 1887 Training 8.38 8.39 7.38 -10.78 FL 1888 Training 8.14 8.74 8.07 -10.37 90 FL 1890 Training 8.45 8.24 8.11 -9.41 FL 1894 Training 8.38 7.97 7.82 -9.25 FL 1896 Training 8.63 8.71 7.52 -11.37 FL 1897 Training, 9.01 7.91 6.93 '-10.78 *FL 1898 Training 8.08 7.75 7.09 -9.74 FL 1900 Validation 8.61 7.94 6.84 -10.77 FL 1903 Validation 9.63 7.96 7.30 -10.64 FL 1904 Validation 8.79 8.14 7.15 -10.82 FL 1905 Validation 8.22 8.24 7.36 -10.43 FL 1906 Training 8.40 7.40 7.24 -8.93 FL 1907 Validation 8.61 7.11 6.59 -9.40 FL 1910 Training 9.47 8.28 7.63 -10.73 FL 1912 Validation 8.32 8.45 8.52 -9.18 FL 1913 Training 8.24 8.60 7.23 -11.41 FL 1916 Validation 8.31 8.04 7.27 -10.19 FL 1918 Validation 8.30 8.49 7.78 -10.37 FL 1919 Training 8.05 8.42 8.00 -9.75 FL 735 Validation 9.03 7.83 7.41 -9.88 FL 738 Validation 9.54 8.07 7.65 -10.30 FL 739 Training 8.14 8.09 7.69 -9.57 FL 878 Validation 9.17 7.91 7.70 -9.69 FL 879 Training 8.37 7.96 7.67 -9.45 FL 886 Validation 8.59 8.38 7.67 -10.44 FL 888 Validation 7.85 7.71 7.07 -9.56 FL 1627 Training 8.26 8.17 7.36 -10.30 FL 1429 Training 9.53 8.12 7.56 -10.53 FL 1850 Validation 9.81 8.00 7.99 -9.77 FL 1735 Validation 8.46 8.10 7.62 -9.86 In order to visualize the predictive power of the model, the FL samples were ranked according to their survival predictor scores and divided into four quartiles. Kaplan-Meier plots of overall survival showed clear differences in survival rate in the 5 validation set (Figure 9). Example 6: Development of a first DLBCL survival predictor using gene expression data from Affymetrix U133A and U133B microarrays: Gene expression data from Affymetrix U133A and U133B microarrays was obtained for 231 DLBCL samples. The follow-up time and status at follow-up for [0 each of the subjects from whom these samples were acquired is listed in Table 13. Table 2368 also indicates which samples were used in creating the survival predictor.

9 1 Table 13 Sample ID # Length of follow- Status at follow-up Used in creating survival up (years) predictor? ABC 1000 0.69 Dead Yes ABC 1002 0.28 Dead Yes ABC 1023 5.57 Dead Yes ABC 1027 0.25 Dead Yes ABC 1031 6.64 Dead Yes ABC 1034 2.31 Dead Yes ABC 1038 0.71 Dead Yes ABC 1043 2.31 Dead Yes ABC 1045 2.26 Dead Yes ABC 1055 7.81 Alive Yes ABC 1057 2.13 Dead Yes ABC 1059 2.00 Dead Yes ABC 1061 1.04 Dead Yes ABC 1946 0.68 Dead No ABC 1994 1.21 Dead No ABC 2001 1.32 Dead No ABC 304 1.31 Dead Yes ABC 305 0.82 Alive Yes ABC 309 2.80 Alive Yes ABC 413 0.60 Dead Yes ABC 428 11.38 Alive Yes ABC 432 0.38 Dead Yes ABC 446 2.82 Dead Yes ABC 462 7.49 Dead Yes ABC 477 1.70 Dead Yes ABC 481 10.75 Alive Yes ABC 482 7.72 Alive Yes ABC 538 0.34 Dead Yes ABC 541 4.11 Alive Yes ABC 544 1.31 Dead Yes ABC 547 0.05 Dead Yes ABC 577 1.65 Alive Yes ABC 616 0.99 Dead Yes ABC 626 2.49 Dead Yes ABC 633 2.02 Alive Yes ABC 642 0.34 Dead Yes ABC 644 0.31 Dead Yes ABC 645 6.08 Dead Yes ABC 646 2.59 Dead Yes ABC 651 2.34 Alive Yes ABC 652 0.01 Dead Yes ABC 660 0.20 Dead Yes ABC 663 0.62 Dead Yes ABC 668 6.44 Alive Yes ABC 676 1.00 Dead Yes ABC 678 0.06 Dead Yes ABC 687 0.94 Dead Yes ABC 689 2.54 Dead Yes ABC 692 10.53 Alive Yes ABC 694 4.83 Alive Yes ABC 700 5.40 Dead Yes ABC 702 4.13 Dead Yes ABC 704 9.67 Alive Yes 92 ABC 709 0.47 Dead Yes ABC 712 3.26 Dead Yes ABC 714 2.45 Dead Yes ABC 717 0.42 Dead Yes ABC 725 0.96 Dead Yes ABC 726 7.62 Alive Yes ABC 730 1.03 Dead Yes ABC 753 0.04 Dead Yes ABC 756 7.21 Alive Yes ABC 771 6.80 Dead Yes ABC 779 0.35 Dead Yes ABC 800 0.33 Dead Yes ABC 807 0.31 Dead Yes ABC 809 0.51 Dead Yes ABC 816 1.86 Dead Yes ABC 820 1.59 Dead Yes ABC 823 0.16 Dead Yes ABC 835 1.22 Dead Yes ABC 839 0.29 Dead Yes ABC 841 10.14 Alive Yes ABC 858 3.58 Dead Yes ABC 872 5.00 Alive Yes ABC 875 8.45 Alive Yes ABC 912 16.79 Alive Yes ABC 996 0.21 Dead Yes GCB 1005 5.77 Alive Yes GCB 1008 6.46 Alive Yes GCB 1009 9.68 Alive Yes GCB 1021 14.59 Alive Yes GCB 1025 2.86 Dead Yes GCB 1026 6.94 Dead Yes GCB 1037 0.23 Dead Yes GCB 1039 2.05 Dead Yes GCB 1049 1.33 Dead Yes GCB 1051 0.12 Dead Yes GCB 1058 0.42 Dead Yes GCB 1060 6.45 Alive Yes GCB 1990 0.06 Dead No GCB 1991 1.01 Dead No GCB 2017 0.08 Dead No GCB 2018 0.17 Dead No GCB 2095 0.97 Alive No GCB 412 12.12 Alive Yes GCB 415 5.38 Dead Yes GCB 421 1.24 Dead Yes GCB 424 10.62 Dead Yes GCB 433 0.76 Dead Yes GCB 434 10.53 Alive Yes GCB 438 8.15 Alive Yes GCB 459 9.65 Alive Yes GCB 470 11.17 Alive Yes GCB 479 7.24 Alive Yes GCB 492 11.29 Alive Yes GCB 517 3.03 Dead Yes GCB 523 8.36 Alive Yes GCB 524 5.88 Alive Yes 93 GCB 529 1.06 Dead Yes GCB 533 0.71 Dead Yes GCB 537 4.99 Dead Yes GCB 543 3.47 Alive Yes GCB 545 1.10 Dead Yes GCB 549 2.68 Dead Yes GCB 550 21.78 Alive Yes GCB 553 0.82 Dead Yes GCB 565 9.11 Dead Yes GCB 572 14.24 Alive Yes GCB 617 5.88 Alive Yes GCB 618 5.65 Alive Yes GCB 619 8.76 Alive Yes GCB 623 2.43 Alive Yes GCB 627 1.27 Dead Yes GCB 654 7.37 Alive Yes GCB 661 0.56 Alive Yes GCB 669 7.11 Alive Yes GCB 672 6.78 Alive Yes GCB 674 7.22 Alive Yes GCB 675 6.02 Alive Yes GCB 681 9.70 Alive Yes GCB 688 0.33 Dead Yes GCB 695 0.15 Dead Yes GCB 698 3.88 Alive Yes GCB 701 3.90 Alive Yes GCB 710 1.08 Dead Yes GCB 711 3.93 Dead Yes GCB 722 3.32 Alive Yes GCB 724 1.40 Dead Yes GCB 731 10.18 Alive Yes GCB 742 4.09 Alive Yes GCB 744 8.86 Alive Yes GCB 745 1.33 Dead Yes GCB 747 15.41 Alive Yes GCB 749 10.40 Alive Yes GCB 758 1.10 Dead Yes GCB 772 2.48 Alive Yes GCB 777 4.27 Dead Yes GCB 792 5.53 Alive Yes GCB 795 3.43 Alive Yes GCB 797 6.87 Dead Yes GCB 803 1.45 Dead Yes GCB 810 11.72 _ Alive Yes GCB 817 2.76 Dead Yes GCB 818 0.10 Dead Yes GCB 819 0.72 Dead Yes GCB 821 9.47 Alive Yes GCB 832 4.01 Alive Yes GCB 836 4.29 Alive Yes GCB 840 3.40 Alive Yes GCB 847 4.16 Alive Yes GCB 860 3.03 Dead Yes GCB 871 0.41 Dead Yes GCB 874 0.12 Dead Yes GCB 995 6.65 Alive Yes 94 PMBL 1006 7.12 Alive Yes PMBL 1024 19.83 Alive Yes PMBL 1048 7.70 Alive Yes PMBL 1053 1.04 Dead Yes PMBL 1920 1.97 Alive No PMBL 1921 4.16 Alive No PMBL 1923 1.60 Alive No PMBL 1924 6.11 Alive No PMBL 1935 12.42 Alive No PMBL 1941 0.71 Alive No PMBL 1942 0.88 Alive No PMBL 1943 8.96 Alive No PMBL 1945 0.84 Dead No PMBL 1948 7.96 Alive No PMBL 1949 4.28 Alive No PMBL 1989 1.33 Dead No PMBL 1992 1.00 Dead No PMBL 1993 1.33 Dead No PMBL 2002 6.62 Alive No PMBL 2019 0.99 Dead No PMBL 2020 2.08 Alive No PMBL 2092 1.27 Alive No PMBL 484 1.40 Dead Yes PMBL 546 0.78 Dead Yes PMBL 570 14.40 Alive Yes PMBL 621 8.14 Alive Yes PMBL 638 0.70 Dead Yes PMBL 691 0.32 Dead Yes PMBL 791 1.33 Dead Yes PMBL 824 12.24 Alive Yes PMBL 906 16.80 Alive Yes PMBL 994 4.79 Alive Yes PMBL 998 9.11 Alive Yes UC DLBCL 1001 0.33 Dead Yes UC DLBCL 1004 6.72 Alive Yes UC DLBCL 1007 2.26 Dead Yes UC DLBCL 1018 0.03 Dead Yes UC DLBCL 1041 3.13 Dead Yes UC DLBCL 1054 12.34 Alive Yes UC DLBCL 306 2.69 Alive Yes UC DLBCL 310 0.97 Alive Yes UC DLBCL 449 9.16 Alive Yes UC DLBCL 452 9.17 Alive Yes UC DLBCL 458 1.18 Dead Yes UC DLBCL 460 9.02 Alive Yes UC DLBCL 491 4.47 Dead Yes UC DLBCL 528 1.64 Alive Yes UC DLBCL 615 4.94 Alive Yes UC DLBCL 625 5.24 Alive Yes UC DLBCL 664 0.62 Dead Yes UC DLBCL 671 3.35 Alive Yes UC DLBCL 682 0.11 'Dead Yes UC DLBCL 683 7.42 Alive Yes UC DLBCL 684 1.92 Dead Yes UC DLBCL 748 1.01 Dead Yes UC DLBCL 751 9.99 Alive Yes 95 UC DLBCL 808 0.37 Dead Yes UC DLBCL 831 11.02 Dead Yes UC DLBCL 834 1.64 Dead Yes UC DLBCL 838 0.00 Dead Yes UC DLBCL 851 0.05 Dead , Yes UC DLBCL 854 1.51 Dead Yes UC DLBCL 855 1.67 Alive Yes UC DLBCL 856 0.60 Dead Yes The correlation between expression of each gene represented on the microarrays and survival was estimated using a Cox proportional hazards model. A positive Cox coefficient indicated increasing mortality with increasing expression of 5 the gene, while a negative Cox coefficient indicated decreasing mortality with increasing expression of the gene. Genes that were significantly correlated with survival (p<0.001) were grouped into gene expression signatures using a hierarchical clustering algorithm. The expression level of every component gene in each of these gene expression 10 signatures was averaged for each sample to create a gene expression signature value. A step-up procedure (Drapner 1966) was applied to determine the optimal number of gene signatures to use in the survival predictor model. First, the gene expression signature that was most significantly associated with survival was included in the model. Next, the gene expression signature with the second highest 15 association with survival was added to the model to form a two-component model. This procedure was repeated until there was no gene expression signature to add to the model with a p-value of <0.05. The final prediction model incorporated gene expression signature values from three gene expression signatures. The first gene expression signature added 20 to the model was termed "ABC DLBCL high," because it included genes that were more highly expressed in ABC than in GCB (Rosenwald 2002). The second gene 96 expression signature added to the model was termed "lymph node," because it reflected the response of non-tumor cells in the lymph node to the malignant lymphoma cells. The final gene expression signature added to the model was termed "MHC class 11," because it included all of the genes encoding the MHC class 5 Il alpha and beta chains. Table 14 shows the genes that were averaged to form each of these signatures. Table 14 Signature UNIQID Gene symbol Survival p-value ABC DLBCL high 1134271 POU5Fi 3.09E-05 ABC DLBCL high 1121564 DRIL1 4.06E-05 ABC DLBCL high 1119889 PDCD4 7.28E-05 ABC DLBCL high 1133300 CTH 1.23E-04 ABC DLBCL high 1106030 MGC:50789 1.70E-04 ABC DLBCL high 1139301 FLJ20150 4.49E-04 ABC DLBCL high 1122131 CHST7 5.18E-04 ABC DLBCL high 1114824 LIMD1 5.20E-04 ABC DLBCL high 1100161 LOC142678 6.24E-04 ABC DLBCL high 1120129 TLE1 6.95E-04 Lymph node 1097126 TEM8 5.14E-09 Lymph node 1120880 LTBP2 9.80E-07 Lymph node 1098898 FLJ31066 1.09E-06 Lymph node 1123376 RARRES2 1.68E-06 Lymph node 1128945 SLC12A8 2.90E-06 Lymph node 1130994 DPYSL3 3.37E-06 Lymph node 1124429 SULF1 3.53E-06 Lymph node 1099358 FLJ39971 4.09E-06 Lymph node 1130509 SPARC 6.23E-06 Lymph node 1095985 TMEPAI 7.07E-06 Lymph node 1123038 ACTN1 7.90E-06 Lymph node 1133700 CDH1 1 8.20E-06 Lymph node 1122101 TFEC 9.66E-06 Lymph node 1124296 SDC2 9.99E-06 MHC Class I 1123127 HLA-DRA 1.21 E-06 MHC Class Il 1136777 HLA-DQAI 3.45E-06 MHC Class I 1137771 HLA-DRB1 3.95E-06 MHC Class I 1134281 HLA-DRB4 2.70E-05 MHC Class Il 1136573 HLA-DPAI 2.92E-05 MHC Class |1 1132710 HLA-DRB3 7.09E-05 Fitting the Cox proportional hazards model to the three gene expression 10 signature values resulted in the following model: Survival predictor score = [0.586*(ABC DLBCL high gene expression signature value)] - [0.468*(lymph node gene 97 expression signature value)] - [0.336*(MHC Class 11 gene expression signature v.alue)]. A higher survival predictor score was associated with worse outcome. According to a likelihood ratio test adjusted for the number of variables included, this model was 5 significant in predicting survival at p=2.13x10' 3 . In order to visualize the predictive power of the model, the 205 samples used to create the model were ranked according to their survival predictor scores and divided into four quartiles. Kaplan Meier plots of overall survival probability show clear differences in survival rate between these four quartiles (Figure 10). The five-year survival probabilities for 0 each quartile are set forth in Table 15. Table 15 Quartile 5-year survival 1 83% 2 59% 3 33% 4 17% Example 7: Development of a second DLBCL survival predictor using gene expression data from the Lymph Dx microarray: 15 A DLBCL survival model based on gene expression had been developed previously using proliferation, germinal center B-cell, lymph node, and MHC class Il gene expression signatures and the expression of the single gene BMP-6 (Rosenwald 2002). BMP-6 expression was poorly measured on the Lymph Dx microarray, but genes associated with each of these four gene expression 2O signatures exhibited associations with survival similar to those observed using Lymphochip microarrays. DLBCL samples were divided into two groups: a training set (100 samples) for developing the survival prediction model, and a validation set (100 samples) for evaluating the reproducibility of the model. Gene expressed in the 98 training set samples were clustered, and lymph node, germinal center B-cell, MHC class 11, and proliferation gene expression signatures were identified. Within each signature, expression of genes that were associated with survival (p<0.01) was averaged to generate a gene expression signature value for each signature. Table 5 16 lists the genes that were used to generate the gene expression signature value for each signature. Table 16 Signature UNIQID Unigene ID Build 167 Gene symbol (http://www.ncbi.nlm. nih.gov/UniGene) Germinal center B-cell 1099686 117721 Germinal center B-cell 1099711 243596 Germinal center B-cell 1103390 271752 BPNT1 Germinal center B-cell 1106025 49500 KIAA0746 Germinal center B-cell 1128287 300063 ASB13 3 Germinal center B-cell 1132520 283063 LMO2 Germinal center B-cell 1138192 126608 NR3C1 Germinal center B-cell 1529318 291954 Germinal center B-cell 1529344 317970 SERPINA1 1 Germinal center B-cell 1529352 446195 Germinal center B-cell 1096570 409813 ANUBL1 Germinal center B-cell 1097897 266175 PAG Germinal center B-cell 1097901 266175 PAG Germinal center B-cell 1098611 433611 PDKI Germinal center B-cell 1100581 155024 BCL6 Germinal center B-cell 1115034 387222 NEK6 Germinal center 3-cell 1120090 155024 BCL6 Germinal center B-cell 1120946 25209 MAPKI0 Germinal center B-cell 1121248 54089 BARD1 Germinal center B-cell 1123105 434281 PTK2 Germinal center B-cell 1125456 300592 MYBLI Germinal center B-cell 1128694 171466 ELL3 Germinal center B-cell 1128787 114611 C7orfl10 Germinal center B-cell 1132122 307734 MME Germinal center B-cell 1136269 101474 MAST2 Germinal center B-cell 1136702 155584 KIAA0121 Germinal center B-cell 1139230 29724 PLEKHF2 Germinal center B-cell 1529292 NA Germinal center B-cell 1529295 116441 Lymph node 1097126 274520 ANTXR1 Lymph node 1099028 334838 FNDC1 Lymph node 1099358 93135 Lymph node 1101478 146246 MGC45780 Lymph node 1103497 50115 Lymph node 1121029 412999 CSTA Lymph node 1124429 409602 SULF1 Lymph node 1135068 71719 PDLIM3 Lymph node 1136051 520937 CSF2RA 99 Lymph node 1136172 38084 SULTIC1 MHC class 11 1136777 387679 HLA-DQA1 MHC class 1i 1136877 409934 HLA-DOB1 Proliferation 1096903 437460 FLJ10385 Proliferation 1120583 153768 RNU31P2 Proliferation 1123289 5409 POLRIC Proliferation 1131808 75447 RALBP1 Proliferation 1133102 360041 FRDA Proliferation 1136595 404814 VDAC1 Table 17 lists p-values for the association of each signature with survival in the training set, the validation set, and overall. Table 17 Signature Training set Validation set Overall Lymph node 4.0 x 10-0 2.3 x 10 6.8 x 101u Proliferation 8.1 x 10~0 3.4 x 103 2.1 x 10 Germinal center B-cell 6.2 x 104 | 2.1 x 10 - 5.0 x 10 MHC class il 2.4 x 10* { 2.7 x 10 3.1 x 104 5 The four gene expression signatures were used to generate a survival predictor score using the following equation: Survival predictor score = [-0.4337*(lymph node gene expression signature value)] + [0.09*(proliferation gene expression 10 signature value)] - [0.4144*(germinal center B cell gene expression signature value)] [0.2006*(MHC class I gene expression signature value)]. A higher survival predictor score was associated with worse outcome. For the 200 15 DLBCL samples used to generate the model, the survival predictor score had a mean of 5.7 and a standard deviation of 0.78, with each unit increase in the predictor score corresponding to an approximately 2.7 fold increase in the relative risk of death. Data for all 200 samples is presented in Table 18. Table 18 100 Sample ID # Set Lymph Proliferation Germinal MHC Survival node signature center B- class il predictor signature value cell signature score value signature value value ABC 1000 Validation 6.50 8.92 7.60 11.50 -5.08 ABC 1002 Validation 7.00 8.58 7.27 12.54 -5.50 ABC 1023 Validation 7.43 8.99 6.80 11.42 -5.05 ABC 1027 Training 5.68 9.00 6.87 12.31 -4.70 ABC 1031 Validation 8.02 9.00 7.17 11.68 -5.53 ABC 1034 Validation 6.06 9.61 6.72 11.83 -4.58 ABC 1038 Training 6.83 8.97 7.17 12.30 -5.23 ABC 1043 Training 6.96 9.01 6.77 12.29 -5.11 ABC 1045 Validation 8.18 8.21 6.77 12.07 -5.66 ABC 1055 Validation 5.58 9.16 7.30 13.05 -4.76 ABC 1057 Training 7.33 8.94 7.74 12.05 -5.53 ABC 1059 Validation 9.02 8.46 7.15 11.35 -6.08 ABC 1061 Training 7.13 9.18 7.09 12.28 -5.21 ABC 304 Validation 5.92 8.80 6.76 12.76 -4.84 ABC 305 Training 5.92 8.74 7.50 11.89 -4.91 ABC 309 Validation 8.86 8.39 7.62 12.53 -6.46 ABC 413 Validation 6.45 9.32 6.55 9.04 -4.16 ABC 428 Training 7.52 9.19 7.98 10.25 -5.51 ABC 432 Validation 6.48 9.33 7.45 9.56 -4.56 ABC 446 Training 7.91 9.42 7.41 10.55 -5.46 ABC 462 Validation 6.41 8.85 6.67 13.36 -5.03 ABC 477 Validation 6.26 9.02 6.69 12.45 -4.89 ABC 481 Training 8.18 8.30 7.35 11.98 -5.91 ABC 482 Training 8.59 9.01 7.66 12.35 -6.16 ABC 538 Validation 8.06 8.84 7.17 11.83 -5.69 ABC 541 Training 6.14 8.52 7.42 10.59 -4.71 ABC 544 Training 6.91 9.03 6.82 11.87 -4.89 ABC 547 Validation 5.80 8.96 7.14 11.38 -4.60 ABC 577 Validation 7.84 8.65 8.16 11.95 -5.94 ABC 616 Validation 6.03 9.05 7.36 .12.64 -4.84 ABC 626 Validation 7.48 9.22 7.25 11.11 -5.27 ABC 633 Training 7.74 8.35 7.39 12.45 -5.80 ABC 642 Training 5.71 8.82 6.41 13.80 -4.62 ABC 644 Validation 6.64 9.15 7.05 13.28 -5.20 ABC 645 Training 8.44 8.81 7.93 13.39 -6.43 ABC 646 Validation 5.94 9.11 6.71 11.60 -4.63 ABC 652 Validation 5.87 8.85 6.88 12.73 -4.77 ABC 660 Training 5.19 9.34 6.64 10.17 -3.86 ABC 663 Training 5.69 9.02 7.33 12.82 -4.91 ABC 668 Validation 7.12 9.28 7.03 10.57 -4.91 ABC 676 Training 4.95 8.90 7.09 13.32 -4.61 ABC 678 Training 5.84 9.11 7.34 11.26 -4.41 ABC 687 Validation 5.15 9.89 6.56 10.46 -3.76 ABC 689 Training 6.49 8.86 7.10 12.56 -4.88 ABC 692 Validation 7.32 8.96 7.25 11.57 -5.32 ABC 694 Validation 8.28 9.21 8.01 12.41 -6.23 ABC 700 Training 7.29 8.97 7.55 12.10 -5.48 ABC 702 Validation 7.60 8.66 6.86 12.55 -5.45 ABC 704 Training 7.07 8.92 7.03 12.83 -5.35 ABC 709 Validation 5.92 8.58 . 6.37 13.40 -4.66 ABC 712 Validation 5.79 9.12 6.34 12.02 -4.23 101 ABC 714 Training 7.49 8.88 7.49 11.97 -5.54 ABC 717 Train 7.17 9.45 7.01 11.3 -5.05 ABC 725 Training 6.71 9.01 6.52 12.76 -4.86 ABC 726 Validation 6.91 8.72 6.71 11.91 4.90 ABC 730 Validation 6.28 9.22 7.28 12.14 -4.88 ABC 753 Traiin 6.84 9.64 7.05 13.00 -5.22 ABC 756 Trann 7.67 8.45 7.59 12.48 -5.85 ABC 771 Trani 6.98 8.76 6.91 12.20 -5.18 ABC 779 Trang 6.73 9.32 6.78 9.82 -4.44 ABC 800 Validation 8.75 8.31 7.45 11.91 -6.04 ABC 807 Training 5.50 9.53 6.92 7.56 -3.79 ABC 809 Training 7.40 8.70 7.68 10.83 -5.50 ABC 816 Training 5.20 9.91 7.65 10.64 -4.14 ABC 820 Training 6.71 8.94 6.55 11.98 4.85 ABC 823 Validation 5.58 9.26 6.44 10.09 -3.97 ABC 835 Validation 6.95 8.68 8.04 12.31 -5.59 ABC 839 Training 6.63 9.17 7.23 11.89 -5.04 ABC 841 Validation 6.35 9.51 7.52 13.19 -5.28 ABC 858 Trainin 7.63 8.51 7.12 11.74 -5.42 ABC 872 Training 6.78 1 8.73 7.41 12.47 -5.44 ABC 875 Training 7.59 8.81 7.20 11.26 -5.25 ABC 912 Validation 7.01 8.55 7.45 12.79 -5.64 ABC 996 Validation 5.00 9.53 6.70 10.02 -3.94 GCB 1005 Validation 8.28 8.67 9.11 13.27 -6.98 GCB 1008 Training 8.17 8.59 9.83 12.83 -7.06 GCB 1009 Training 6.63 9.02 10.07 12.28 -6.19 GOB 1021 Validation 6.44 8.83 9.34 13.20 -6.15 GCB 1025 Validation 7.87 8.48 9.27 12.37 -6.57 GCB 1026 Training 7.71 8.30 9.81 13.52 -6.85 GCB 1037 Training 4.95 8.83 9.35 12.57 -5.22 GCB 1039 Training 7.63 8.65 9.01 13.28 -6.47 GCB 1049 Validation 8.54 8.61 8.12 12.60 -6.41 GOB 1051 Validation 6.26 9.09 9.48 12.76 5*97 GOB 1058 Validation 7.12 8.89 8.34 12.80 -5.85 GOB 1060 Validation 8.27 8.84 8.94 12.96 -6.75 GOB 412 Training 7.22 8.33 8.50 13.09 -6.09 GOB 415 Training 9.01 8.62 8.38 11.99 -6.47 GCB 421 Training 7.59 7.89 7.49 12.20 -5.80 GCB 424 Training 9.29 8.42 8.51 12.44 -6.79 GOB 433 Training 8.45 8.34 8.02 12.64 -6.54 GOB 434 Training 8.46 8.55 9.17 12.54 -6.98 GCB 438 Validation 8.14 8.71 9.13 12.51 -6.67 GOB 459 Validation 8.98 8.39 8.42 11.37 -6.49 GOB 470 Validation 7.72 8.57 8.67 12.23 -6.12 GOB 479 Validation 6.86 _8.25 7.-3 13.07 -5.35 GOB 492 Training 8.01 8.61 9.51 12.34 -6.63 GOB 517 Validation 8.57 8.73 7.99 12.76 -6.48 GCB 523 Trainin 5.96 8.56 8.74 12.77 -5.72 GOB 524 Training 8.51 8.09 8.76 12.51 -6.57 GOB 529 Training 5.12 9.17 8.88 10.77 -4.86 GOB 533 Training 8.88 8.81 8.36 12.44 -6.60 GOB 537 Validation 7.42 8.19 9.73 13.29 -6.68 GOB 543 Validation 8.49 8.02 8.66 12.06 -6.45 GOB 545 Training 8.65 8.28 6.90 12.90 -6.13 GOB 549 Validation 6.87 8.24 8.65 12.15 -6.00 GOB 550 Validation 8.98 1 8.29 8.76 12.24 -6.94 102 GCB 553 Validation 8.51 8.64 8.62 12.63 -6.69 GCB 565 Validation 7.97 8.79 . 9.79 13.42 -6.98 GCB 572 Training 7.61 8.60 9.39 12.58 -6.42 GCB 617 Validation 8.31 7.89 7.54 13.17 -6.12 GCB 618 Training 5.66 8.97 9.20 13.32 -5.54 GCB 619 Validation 7.83 8.65 9.34 12.12 -6.36 GCB 623 Training 7.16 8.88 9.26 12.35 -6.21 GCB 627 Validation 8.13 8.83 8.62 11.85 -6.31 GCB 654 Training 6.30 9.60 8.45 10.00 -4.88 GCB 661 Validation 8.46 8.51 8.18 12.66 -6.33 GCB 669 Training 7.88 8.65 8.59 12.32 -6.19 GCB 672 Training 8.29 8.61 8.14 12.41 -6.21 GCB 674 Validation 8.36 8.62 7.76 12.33 -6.14 GCB 675 Validation 6.01 9.52 8.90 10.12 -5.09 GCB 681 Training 9.25 8.72 8.72 12.59 -6.89 GCB 688 Validation 6.97 9.01 9.90 9.94 -5.99 GCB 695 Validation 8.80 8.73 9.23 12.45 -6.84 GCB 698 Validation 9.27 8.35 8.85 11.99 -6.96 GCB 701 Training 7.77 7.93 8.68 13.10 -6.33 GCB 710 Validation 6.12 8.78 7.65 13.19 -5.24 GCB 711 Training 7.57 8.80 8.43 11.44 -5.84 GCB 722 Training 7.78 8.31 8.93 12.61 -6.51 GCB 724 Training 7.88 9.08 8.74 11.53 -6.21 GCB 731 Validation 7.72 8.92 9.08 12.20 -6.46 GCB 742 Validation 8.33 8.55 8.58 12.95 -6.70 GCB 744 Training 8.02 8.64 9.36 11.85 -6.52 GCB 745 Training 8.47 8.34 8.93 11.95 -6.67 GCB 747 Validation 7.64 8.48 8.32 13.06 -6.27 GCB 749 Training 7.57 8.61 9.40 12.55 -6.56 GCB 758 Validation 5.66 8.77 7.89 12.51 -4.63 GCB 772 Validation 8.52 7.81 7.95 12.25 -6.34 GCB 777 Validation 7.52 8.65 8.57 11.69 -6.10 GCB 792 Training 8.14 8.64 9.21 12.08 -6.65 GCB 795 Validation 9.19 8.17 8.81 11.60 -6.92 GCB 797 Validation 7.50 8.62 8.08 12.84 -6.09 GCB 803 Validation 6.19 8.65 9.49 13.18 -6.11 GCB 810 Training 8.46 8.32 8.10 13.13 -6.50 GCB 817 Training 6.93 8.51 9.49 11.09 -6.04 GCB 818 Training 7.18 8.96 8.08 12.23 -5.76 GCB 819 Validation 7.16 8.97 8.06 13.22 -5.79 GCB 821 Validation 8.13 8.59 8.90 12.41 -6.61 GCB 832 Training 7.83 8.35 8.71 12.47 -6.37 GCB 836 Validation 7.84 8.99 8.50 11.46 -5.85 GCB 840 Training 8.24 7.75 7.40 11.74 -5.77 GCB 847 Training 7.82 8.17 8.97 12.55 -6.51 GCB 860 Training 7.12 8.39 9.34 11.54 -6.10 GCB 871 Training 5.59 9.60 7.28 11.16 -4.23 GCB 874 Training 8.53 9.14 8.95 11.65 -6.47 GCB 995 Validation 6.98 8.68 8.54 12.22 -5.76 PMBL 1006 Validation 7.34 8.51 7.66 10.94 -5.33 PMBL 1024 Validation 7.62 8.48 8.56 10.89 -5.96 PMBL 1048 Validation 8.68 8.16 7.23 12.18 -6.08 PMBL 1053 Training 7.02 8.28 8.24 11.12 -5.31 PMBL 484 Training 7.15 8.45 7.01 13.62 -5.41 PMBL 546 Validation 8.19 7.88 7.66 11.73 -6.06 PMBL 570 Training 9.34 8.21 8.48 12.70 -6.86 103 PMBL 621 Training 8.08 8.60 9.14 12.96 -6.72 PMBL 638 Training 7.56 8.26 8.00 11.37 -5.75 PMBL 691 Validation 6.48 8.92 8.40 10.17 -5.04 PMBL 791 Validation 7.72 8.65 8.94 11.56 -6.16 PMBL 824 Validation 8.06 8.01 7.76 13.28 -6.11 PMBL 994 Training 9.15 8.36 7.46 12.43 -6.29 PMBL 998 Training 6.70 8.35 9.24 13.19 -6.20 UC DLBCL 1001 Validation 6.74 8.43 7.10 12.76 -5.31 UC DLBCL 1004 Validation 7.54 8.75 8.01 13.09 -6.10 UC DLBCL 1007 Training 9.97 8.44 7.64 12.97 -6.85 UC DLBCL 1018 Training 6.42 8.38 6.97 12.71 -5.03 UC DLBCL 1041 Validation 5.76 8.69 6.78 13.38 -4.71 UC DLBCL 1054 Training 8.92 8.65 8.51 11.48 -6.59 UC DLBCL 306 Validation 7.85 8.90 8.31 12.36 -6.23 UC DLBCL 310 Training 8.14 8.80 7.63 12.27 -6.03 UC DLBCL 449 Validation 9.03 8.48 7.07 12.17 -6.01 UC DLBCL 458 Training 5.92 8.53 8.28 9.60 -4.96 UC DLBCL 460 Validation 7.92 9.08 8.30 12.29 -6.13 UC DLBCL 491 Training 7.65 8.33 7.35 12.39 -5.53 UC DLBCL 528 Validation 6.99 8.56 7.36 11.63 -5.35 UC DLBCL 615 Validation 7.11 8.32 8.77 12.80 -6.10 UC DLBCL 625 Training 8.93 7.78 7.85 12.62 -6.46 UC DLBCL 664 Training 7.62 8.15 8.17 12.72 -6.04 UC DLBCL 671 Training 8.09 8.48 7.61 11.53 -5.78 UC DLBCL 682 Training 7.38 8.35 7.14 12.33 -5.43 UC DLBCL 683 Training 7.91 8.36 7.78 12.57 -6.02 UC DLBCL 684 Validation 8.06 8.63 8.29 12.76 -6.29 UC DLBCL 748 Validation 5.38 8.57 7.45 9.55 -4.23 UC DLBCL 751 Training 6.33 8.65 8.88 13.14 -5.74 UC DLBCL 808 Training 7.42 9.01 7.44 13.09 -5.63 UC DLBCL 831 Validation 8.33 8.30 7.46 11.58 -5.84 UC DLBCL 834 Training 6.98 9.09 8.61 11.77 -5.66 UC DLBCL 830 Validation 7.25 8.40 7.23 12.56 -5.36 UC DLBCL 851 Validation 6.28 9.05 6.78 8.19 -4.10 UC DLBCL 854 Validation 7.36 8.50 7.39 12.59 -5.53 UC DLBCL 855 Training 8.31 7.94 7.49 12.08 -6.07 UC DLBCL 856 Validation 5.65 9.01 8.52 9.32 -4.68 In order to visualize the predictive power of the model, the 200 samples were ranked according to their survival predictor scores and divided into four quartiles. Kaplan-Meier plots of overall survival probability show clear differences in survival 5 rate between these four quartiles (Figure 11). Example 8: Development of a third DLBCL survival predictor using gene expression data from the Lymph Dx microarray: The number of genes used to generate the DLBCL survival predictor in Example 7 were reduced in order to create a survival predictor compatible with RT- 1 04 PCR. The list of genes from the lymph node and germinal center B-cell gene expression signatures was narrowed to those three genes from.each signature that were most closely correlated with the lymph node and germinal center B-cell gene expression signature values, respectively. The genes from the proliferation gene 5 expression signature did not add significantly to the reduced gene survival prediction model, so they were removed entirely. The expression of the genes within each signature was averaged on the log2 scale to generate a gene expression signature value for each signature. Table 19 lists the genes that were used to generate these gene expression signature values. 0 Table 19 Signature UNIQID Unigene ID Build Gene symbol 167 http://www.ncbi.nim .nih.gov/UnIGene Germinal center B-cell 1099686 117721 Germinal center B-cell 1529318 291954 Germinal center B-cell 1529344 317970 SERPINAI1 Ly hnode 1097126 274520 ANTXR1 Lymph node 1099358 93135 Lymph node 1121029 412999 cSTA MHC class 11 1136777 387679 HLA-DQA1 MHC class 11 1136877 409934 HLA-DQB1 Table 20 lists p-values for the association of each signature with survival in the training set, the validation set, and overall. Table 20 Signature Training set Validation set Overall Lymph node 6.1 x 10 0.0021 2.1 x 10~ Germinal center B-cell 3.5 x 10' 0.0099 2.7 x 10 MHC class 1I 0.024 0.0026 0.00031 5 The three gene expression signatures were used to generate a survival predictor score using the following equation: Survival predictor score = [-0.32*(lymph node gene expression signature value)] - [0.176*(germinal center B-cell gene 105 expression signature value)] - [0.206*(MHC class 11 gene expression signature value)]. . A higher survival predictor score was associated with worse outcome. For the 200 DLBCL samples used to generate the model, the survival predictor score had a 5 mean of 6.54 and a standard deviation of 0.69, with each unit increase in the , predictor score corresponding to an approximately 2.7 fold increase in the relative risk of death. Data for all 200 samples is presented in Table 21. T a b l e 2 1 _ _ _ _-l a s Sample ID # Set Lymph Germinal MHC class Survival node center B-cell If predictor signature signature signature score value value value ABC 1000 Validation 8.08 5.68 11.50 -5.96 ABC 1002 Validation 8.32 6.06 12.54 -6.31 ABC 1023 Validation 9.36 4.74 11.42 -6.18 ABC 1027 Training 7.41 4.90 12.31 -5.77 ABC 1031 Validation 9.40 5.23 11.68 -6.33 ABC 1034 Validation 7.47 4.92 11.83 -5.69 ABC 1038 Training 7.89 5.84 12.30 -6.09 ABC 1043 Training 7.84 4.66 12.29 -5.86 ABC 1045 Validation 9.31 4.66 12.07 ' -6.29 ABC 1055 Validation 6.46 6.38 13.05 -5.88 ABC 1057 Training 9.13 7.93 12.05 -6.80 ABC 1059 Validation 10.93 4.82 11.35 -6.68 ABC 1061 Training 8.18 5.04 12.28 -6.04 ABC 304 Validation 7.31 6.47 12.76 -6.10 ABC 305 Training 7.02 6.60 11.89 -5.86 ABC 309 Validation 10.47 7.00 12.53 -7.16 ABC 413 Validation 7.99 4.80 9.04 -5.26 ABC 428 Training 9.43 7.59 10.25 -6.47 ABC 432 Validation 7.29 8.16 9.56 -5.74 ABC 446 Training 9.49 5.46 10.55 -6.17 ABC 462 Validation 7.72 4.97 13.36 -6.10 ABC 477 Validation 7.16 3.69 12.45 -5.51 ABC 481 Training 9.75 6.89 11.98 -6.80 ABC 482 Training 10.51 7.64 12.35 -7.25 ABC 538 Validation 8.79 5.00 11.83 -6.13 ABC 541 Training 7.70 5.80 10.59 -5.67 ABC 544 Training 8.90 3.98 11.87 -5.99 ABC 547 Validation 7.05 5.18 11.38 -5.51 ABC 577 Validation 9.93 8.05 11.95 -7.06 ABC 616 Validation 7.34 4.54 12.64 -5.75 ABC 626 Validation 8.78 6.77 11.11 -6.29 ABC 633 Training 9.63 5.02 12.45 -6.53 ABC 642 Training 7.31 4.95 13.80 -6.05 ABC 644 Validation 7.72 5.35 13.28 -6.15 ABC 645 Training 9.77 6.21 13.39 -6.98 106 ABC 646 Validation 7.39 3.75 11.60 -5.41 ABC 652 Validation 7.51 4.53 12.73 -5.82 ABC 660 Training 5.85 3.55 10.17 -4.59 ABC 663 Training 7.04 5.06 12.82 -5.78 ABC 668 Validation 8.00 5.65 10.57 -5.73 ABC 676 Training 6.53 4.29 13.32 -5.59 ABC 678 Trainin 6.87 7.48 11.26 -5.83 ABC 687 Validation 6.39 3.78 10.46 -4.87 ABC 689 Training 8.29 5.07 12.56 -6.13 ABC 692 Validation 8.10 5.26 11.57 -5.90 ABC 694 Validation 9.67 8.15 12.41 -7.09 ABC 700 Training 8.37 6.75 12.10 -6.36 ABC 702 Validation 8.44 4.59 12.55 -6.09 ABC 704 Training 8.51 4.34 12.83 -6.13 ABC 709 Validation 7.47 4.54 13.40 -5.95 ABC 712 Validation 7.12 3.99 12.02 -5.46 ABC 714- Trainlng 9.57 7.03 11.97 -6.77 ABC 717 Training 8.33 5.54 11.34 -5.98 ABC 725 Training 8.04 4.40 12.76 -5.97 ABC 726 Validation 7.79 4.18 11.91 -5.68 ABC 730 Validation 8.13 7.36 12.14 -6.40 ABC 753 Training 9.24 6.60 13.00 -6.80 ABC 756 Training 9.51 5.21 12.48 -6.53 ABC 771 Training 8.08 4.74 12.20 -5.93 ABC 779 Training 8.11 4.09 9.82 -5.34 ABC 800 Validation 10.34 4.83 11.91 -6.61 ABC 807 Training 6.58 4.44 7.56 -4.44 ABC 809 Training 9.29 5.72 _ 10.83 -6.21 ABC 816 Training 6.36 6.36 10.64 -5.35 ABC 820 Training 8.10 4.79 11.98 -5.90 ABC 823 Validation 6.63 4.85 10.09 -5.05 ABC 835 Validation 9.17 7.78 12.31 -6.84 ABC 839 Training 8.06 4.97 11.89 -5.90 ABC 841 Validation 8.05 6.24 13.19 -6.39 ABC 858 -Training 9.02 4.86 11.74 -6.16 ABC 872 Training 8.67 5.85 12.47 -6.37 ABC 875 Training 9.60 5.59 11.26 -6.37 ABC 912 Validation 7.99 7.74 12.79 -6.56 ABC 996 Validation 6.89 6.23 10.02 -5.36 GCB 1005 Validation 9.02 9.56 13.27 -7.30 GCB 1008 Training 9.27 10.49 12.83 -7.46 GCB 1009 Training 7.80 10.09 12.28 -6.80 GCB 1021, Validation 8.73 9.20 13.20 -7.13 GCB 1025 Validation 9.94 9.97 12.37 -7.49 GOB 1026 Training 9.54 10.20 13.52 -7.63 GCB 1037 Training 6.34 8.79 12.57 -6.17 GCB 1039 Training 8.71 9.94 13.28 -7.27 GCB 1049 Validation 10.53 8.18 12.60 -7.41 GCB 1051 Validation 7.63 10.18 12.76 -6.86 GCB 1058 Validlation 8.61 9.04 12.80 -6.98 GCB 1060 Validation 10.23 9.38 12-96 -7.59 GCB 412 Training 8.79 7.92 -13.09 -6.90 GCB 415 Training 10.72 8.57 11.99 -7.41 GOB 421 Training 9.23 .5.26 12.20 -6.39 GCB 424 Training 11.14 8.46 12.44 -7.62 GOB 433 Trainng 9.26 8.52 12.64 _-7.07 107 GOB 434 Training 9.73 10.13 12.54 -7.48 GOB 438 Validation 9.60 9.99 12.51 -7.41 GOB 459 Validation 10.51 7.75 11.37 -7.07 GOB 470 Validation 9.56 6.63 12.23 -6.74 GOB 479 Validation 7.77 4.71 13.07 -6.01 GOB 492 Training 8.82 9.52 12.34 -7.04 GOB 517 Validation 9.92 6.96 12.76 -7.03 GOB 523 Training 6.59 9.17 12.77 -6.35 GOB 524 Training 10.00 7.83 12.51 -7.16 GOB 529 Training 5.61 7.93 10.77 -5.41 GOB 533 Training 9.55 5.54 12.44 -6.59 GOB 537 Validation 8.25 10.25 1-3-.29 - -7.18 GOB 543 Validation 9.92 8.85 1i2.06 -7.21 GOB 545 Training 9.69 4.91 12.90 -6.62 GOB _549 Validation 7.86 8.88 12.15 -6.58 GOB 550 Validation 10.64 9.53 12.24 -7.60 GOB 553 Validation 10.14 9.05 12.63 -7.44 GOB 565 Validation 9.08 10.80 13.42 -7.57 GOB 572 Training 8.93 10.03 12.58 -7.21 GOB 617 Validation 9.27 7.80 13.17 -7.05 GOB 618 Training 7.23 9.11 13.32 -6.66 GOB 619 Validation 9.63 9.63 12.12 -7.27 GOB 623 Training 8.94 9.07 12.35 -7.00 GOB 627 Validation 9.72 8.33 11.85 -7.02 GOB 654 Training 7.04 5.60 10.00 -5.30 GOB 661 Validation 10.27 7.92 12.66 -7.29 GOB 669 Training 9.15 9.29 12.32 -7.10 GOB 672 Training 9.69 7.36 12.41 -6.95 GOB 674 Validation 9.93 6.23 12.33 -6.81 GOB 675 Validation 7.48 8.46 10.12 -5.97 GOB 681 Training 10.77 9.52 12.59 -7.72 GOB 688 Validation 8.01 10.17 9.94 -6.40 GOB 695 Validation 10.58 9.38 12.45 -7.60 GOB 698 Validation 10.44 9.00 .11.99 -7.39 GOB 701 Training 9.38 9.27 13.10 -7.33 GOB 710 Validation 6.96 13.19 -5.93 GOB 711 Training 9.28 8.49 11.44 -6.82 GOB 722 Training 8.93 9.51 12.61 -7.13 GOB 724 Training 9.51 8.39 11.53 -6.90 GOB 731 Validation 8.82 9.19 12.20 -6.95 GOB 742 Validation 9.95 9.37 12.95 -7.50 GOB 744 Training 10.23 10.11 11.85 -7.49 GOB 745 Trainin 10.29 9.71 11.95 -7.46 GOB 747 Validation 9.83 9.79 13.06 -7.56 GOB 749 Traing 8.57 10.27 12.55 -7.14 GOB 758 Validation 6.88 5.69 12.51 -5.8 GOB 772 Validation 9.92 7.28 . 12.25 -6.98 GOB 777 Validation 9.03 9.63 11.69 -6.99 GOB 792 Training 9.49 9.06 12.08 -7.12 GOB 795 Validation 11.12 9.02 11.60 -7.54 GOB 797 Validation 8.42 5.90 12.84 -6.38 GOB 803 Validation 7.33 10.11 13.18 -6.84 GOB 810 Training 10.00 8.22 13.13 -7.35 GOB 817 Training 8.60 10.16 11.09 -6.82 GOB 818 Training 9.14 7.78 12.23 -6.81 GOB 819 Validation 9.08. 8.63 13.22 -7.15 108 GOB 821 Validation 10.05 9.81 12.41 -7.50 GCB 832 Training 8.83 6.91 12.47 -6.61 GOB 836 Validation 9.49 7.86 11.46 -6.78 GOB 840 -Training 9.45 5.02 11.74 -6.33 GOB 847 Training 9.41 8.77 12.55 -7.14 GOB 860 Training 9.02 6.66 11.54 -6.43 GOB 871 Training 6.60 4.46 11.16 -5.20 GOB 874 Training 10.39 9.13 11.65 -7.33 GOB 995 Validation 8.52 9.35 12.22 -6.89 PMBL 1006 Validation 8.72 4.67 10.94 -5.86 PMBL 1024 Validation 9.30 8.47 10.89 -6.71 PMBL 1048 Validation 10.30 4.98 12.18 -6.68 PMBL 1053 Training 8.75 9.78 11.12 -6.81 PMBL 484 Training 8.25 4.96 13.62 -6.32 PMBL 546 Validation 9.66 6.07 11.73 -6.57 PMBL 570 Training 10.58 8.54 12.70 -7.50 PMBL 621 Training 9.39 9.94 12.96 -7.43 PMBL 638 Training 9.81 8.35 11.37 -6.95 PMBL 691 Validation 8.37 7.51 10.17 -6.10 PMBL 791 Validation 9.29 8.65 11.56 -6.88 PMBL 824 Validation 9.87 7.19 13.28 -7.16 PMBL 994 -Training 11.27 6.73 12.43 -7.35 PMBL 998 Training 7.92 8.34 13.19 -6.72 UC OLBOL 1001 Validation 8.25 5.63 12.76 -6.26 UC DLBCL 1004 Validaion 9.01 7.01 13.09 -6.81 UC OLBOL 1007 Training 11.42 6.73 12.97 -7.51 UO OLBOL 1018 Training 7.77 4.58 12.71 -5.91 UC OLBOL 1041 Validation 7.90 4.33 13.38 -6.05 UC DLBCL 1054 Training 10.41 8.72 11.48 -7.23 UC DLBOL 306 Validation 9.42 6.54 12.36 -6.71 UC DLBCL 310 Training 9.97 5.50 12.27 -6.69 UIC DLBCL 449 Validation 10.01 5.37 12.17 -6.65 UC OLBOL 458 Training 7.50 5.79 9.60 -5.40 UC OLB9L 460 Validation 10.26 8.27 12.29 -7.27 UC DLBCL 491 Training 9.43 4.73 12.39 -6.40 UC OLBOL 528 Validation 8.42 6.19 11.63 -6.18 UC DLBCL 615 Validation 8.44 9.01 12.80 -6.92 UC DLBCL 625 Tralning 10.43 8.27 12.62 -7.39 UC DLBCL 664 Training 9.80 8.74 12.72 -7.29 UC DLBCL -671 Training 9.42 5.26 11.53 -6.32 UC OLB -L 682 Training 9.01 4.73 12.33 -6.26 UC OLBOL 683 Training 8.85 8.23 12.57 -6.87 UC DLBCL 684 Validlation 9.62 8.78 12.76 -7.25 UC DLBCL 748 Validation 7.60 5.79 9.55 -5.42 UC DLBCL 751 Training 6.40 9.91 13.14 -6.50 UC DLBCL 808 Training 9.44 7.01 13.09 -6.9 UC DLBCL 831 Validation 9.45 5.81 11.58 -6.43 UC OLBOL 834 Training 8.52 7.66 11.77 -6.50 IJC 1JLBCL 838 Validation 8.49 4.60 12.56 -- 6.11 UC OLBOL 851 Validation 7.50 4.82 8.19 -4.94 UC DLBOL 854 Validation 8.35 5.82 12.59 -6.29 UC-DLBCL 855 Training 9.56 5.44 12.08 -6.51 UIC DLBCL 856 Validation 1 6.81 7.49 .9.32 -5.42 109 In order to visualize the predictive power of the model, the 200 samples were ranked according to their survival predictor scores and divided into four quartiles. Kaplan-Meier plots of overall survival probability show clear differences in survival rate between these four quartiles (Figure 12). 5 Example 9: Development of an MCL survival predictor using gene expression data from Affymetrix U133A and U133B microarrays: The connection between higher expression of proliferation genes and worse survival in MCL had previously been documented and validated (Rosenwald 2003). A cluster of proliferation genes had been identified in the DLBCL samples used to 0 create the DLBCL survival predictor described in Example 7. By averaging the expression of these genes, a proliferation gene expression signature value had been developed for the DLBCL samples. The correlation of this signature with each probe set on the U133A and U133B microarrays was determined, and the 22 genes for which the correlation was greater than 0.5 were labeled proliferation genes. The 5 correlation between expression of these proliferation genes and survival in 21 MCL samples was estimated using the Cox proportional hazards model. Table 22 lists these 21 MCL samples. Table 22 Sample ID # Length of follow-up Status at follow-up Used in creating (years) survival predictor? MCL 1012 3.19 Alive Yes MCL 1091 3.03 Alive Yes MCL 1114 0.59 Dead Yes MCL 1128 0.43 Dead Yes MCL 1150 3.21 Dead Yes MCL 1162 0.78 Alive Yes MCL 1166 0.53 Dead Yes MCL 1194 0.55 Alive Yes MCL 885 1.19 Alive Yes MCL 918 1.95 Dead Yes MCL 924 5.48 Dead Yes MCL 925 7.23 Alive Yes MCL 926 5.18 Dead Yes MCL 936 2.80 Alive Yes 110 MCL 939 1.07 Dead Yes MCL 953 2.31 Dead Yes MCL 956 1.40 Dead Yes MCL 964 0.75 Alive Yes MCL 966 0.21 Dead Yes MCL 968 1.59 Dead Yes MCL 970 5.02 Dead Yes Out of the 22 proliferation genes, 11 were significant at a 0.001 level. The expression level of these 11 genes in each of the 21 MCL samples was averaged to generate a proliferation gene expression signature value. No other genes 5 represented on the U133A or U133B microarrays correlated with MCL survival to an extent greater than would be expected by chance, so the final model included only proliferation genes. The 11 genes used to generate the model are presented in Table 23. Table 23 Signature UNIQID Gene Symbol Proliferation 1097290 CIRHIA Proliferation 1101295 FLJ40629 Proliferation 1119729 TK1 Proliferation 1120153 LMNB1 Proliferation 1120494 CDC6 Proliferation 1124745 KIAA0056 Proliferation 1126148 DKFZp586E1 120 Proliferation 1130618 TPli Proliferation 1134753 WHSC1 Proliferation 1139654 ECT2 Proliferation 1140632 IMAGE:52707 0 A survival predictor score for MCL was generated using the following equation: Survival predictor score = 1.66*(proliferation gene expression signature value). 5 This model was associated with survival in a statistically significant manner (p = 0.00018). To illustrate the significance of the model in predicting survival, the 21 MCL samples were divided into two equivalent groups based on their survival predictor scores. Those samples with survival predictor scores above the median were placed in the high proliferation group, while those with survival predictor scores below the median were placed in the low proliferation group. Figure 13 illustrates the Kaplan Meier survival estimates for these two groups. Median survival for the 5 high proliferation group was 1.07 years, while median survival for the low proliferation group was 5.18 years. Example 10: Development of an MCL survival predictor using gene expression data from the Lymph Dx microarray: A set of 21 genes associated with proliferation and poor prognosis in MCL 10 had been identified previously (Rosenwald 2003). Of these 21 genes, only four were represented on the Lymph Dx microarray. In order to find a larger set of genes on the Lymph Dx microarray associated with survival in MCL, Lymphochip expression data (Rosenwald 2003) was re-analyzed and another set of proliferation genes whose expression levels were correlated with poor survival in MCL were identified. [5 Thirteen of these genes were represented on the Lymph Dx microarray (median expression >6 on log2 scale). These 13 genes are listed in Table 24. Table 24 Signature UNIQID Unigene ID Build 167 Gene symbol http://www.ncbi.nim. nih.gov/UniGene Proliferation 1119294 156346 TOP2A Proliferation 1119729 164457 TK1 Proliferation 1120153 89497 LMNB1 Proliferation 1121276 24529 CHEKI Proliferation 1123358 442658 AURKB Proliferation 1124178 446579 HSPCA Proliferation 1124563 249441 WEEI Proliferation 1130799 233952 PSMA7 Proliferation 1131274 374378 CKS1B Proliferation 1131778 396393 UBE2S Proliferation 1132449 250822 STK6 Proliferation 1135229 367676 DUT Proliferation 1136585 80976 MK167 1 12 The expression levels of the 13 genes listed in Table 24 on the Lymph Dx microarray were transformed into the l09 2 scale and averaged to form a proliferation gene expression signature value. This was used to generate a survival predictor score using the following equation: 5 Survival predictor score = 1.66*(proliferation gene expression signature value) For the 21 MCL samples analyzed, the survival predictor score had a mean of 14.85 and a standard deviation of 1.13. Even in this limited sample set, the survival predictor score was significantly associated with prognosis (p=0.0049), with each unit increase in the score corresponding to a 2.7 fold increase in the relative risk of [0 death. Data for all 21 samples is shown in Table 25. Table 25 Sample ID # Proliferation Survival predictor signature value score MCL 1012 8.83 14.658 MCL 1091 8.81 14.625 MCL 1114 10.39 17.247 MCL 1128 10.12 16.799 MCL 1150 8.33 13.828 MCL 1162 8.15 13.529 MCL 1166 9.40 15.604 MCL 1194 7.44 12.350 MCL 885 8.68 14.409 MCL 918 9.33 15.488 MCL 924 8.35 13.861 MCL 925 8.86 14.708 MCL 926 8.14 13.512 MCL 936 8.56 14.21 MCL 939 9.14 15.172 MCL 953 9.25 15.355 MCL 956 9.35 15.521 MCL 964 9.74 16.168 MCL 966 8.76 14.542 MCL 968 9.10 15.106 MCL 970 9.27 15.388 To illustrate the significance of the model in predicting survival, the 21 MCL samples were divided into two equivalent groups based on their survival predictor 5 scores. Those samples with survival predictor scores above the median were 1 13 placed in the high proliferation group, while those with survival predictor scores below the median were placed in the low proliferation group. Figure 14 illustrates the Kaplan Meier survival estimates for these two groups. Example 11: Identification of lymphoma samples as MCL based on Bayesian 5 analysis of gene expression data from Affymetrix UI 33A and U1 33B microarrays: A statistical method based on Bayesian analysis was developed to distinguish MCL samples from samples belonging to other lymphoma types based on gene expression profiling. This method was developed using the gene expression data 0 obtained in Example I forthe following lymphoma types: ABC, GCB, PMBL, BL, FH, FL, MALT, MCL, PTLD, SLL, and splenic marginal zone lymphoma (splenic). To determine the lymphoma type of a sample, a series of predictor models are generated. Each predictor model calculates the probability that the sample belongs to a first lymphoma type rather than a second lymphoma type. A method 5 was developed to determine whether a sample was MCL, or one of the following lymphoma types: ABC, BL, FH, FL, GCB, MALT, PMBL, PTLD, SLL, or splenic. This method required ten different predictor models, each designed to determine whether the sample belonged to MCL or one of the other ten lymphoma types (e.g., MCL vs. ABC, MCL vs. BL, etc.). 0 Several of the lymphoma samples analyzed displayed a tendency towards elevated or reduced expression of genes from the lymph node and proliferation gene expression signatures. These genes are likely to be highly differentially expressed between the lymphoma types, but they do not serve as good predictor genes because they are often variably expressed within a single lymphoma type. For this 1 14 reason, any gene that displayed a correlation with the proliferation or lymph node signatures was eliminated from consideration. For each lymphoma type pair (e.g., MCL vs. ABC, MCL vs. FL, etc.), 20 genes were identified that exhibited the greatest difference in expression between 5 MCL and the second lymphoma type according to a Student's t-test. The choice to use 20 genes was arbitrary. For each sample X, the 20 genes were used to generate a linear predictor score (LPS) according to the following formula: 20 LPS(X) = EtjX, j=1 where X is the expression of gene j in sample X and t is the t-statistic for the 0 difference in expression of gene j between a first lymphoma type and a second lymphoma type. This is merely one method for generating an LPS. Others methods include linear discriminant analysis (Dudoit 2002), support vector machines (Furey 2000), or shrunken centroids (Tibshirani 2002). In addition, there is no requirement that a t-statistic be used as the scaling factor. 5 After an LPS had been formulated for each lymphoma sample, the mean and standard deviation of these LPS's was calculated for each lymphoma type. For a new sample X, Bayes' rule can be used to estimate the probability that the sample belongs to a first lymphoma type rather than a second lymphoma type (Figure 15). In this example, Bayes' rule was used to calculate the probability q that sample X 0 was MCL rather than a second lymphoma type using the following equation: q(X is type 1) f(LPS(X);A,,ci,) #(LPS(X); A, 6,) + #(LPS(X); 142,2) where type I is MCL, type 2 is one of the other nine lymphoma types, #(x; P, a) is the normal density function with mean p and standard deviation o-, A, and 6, are the 115 sample mean and variance of the LPS values for lymphoma type 2. This method was used to develop ten predictor models, one for each pairing of MCL and a second lymphoma type. A sample was classified as MCL if each of the ten predictors generated at least a 90% probability that the sample was MCL. If any of the ten 5 predictors indicated a probability of less than 90%, the sample was classified as non MCL. The 10 sets of 20 genes that were included in these models and the t-statistics for each gene are presented in Tables 26-35. Table 26: MCL vs. ABC predictor genes Gene GenBank Affymetrix Scale UNIQID symbol gi Accession probe set ID Factor 1103711 10433184 AK021895.1 232478 at 17.88496416 1133111 PDE9A 48762717 NM 001001567.1 205593 s at 17.61579873 1137987 PLXNBI 41152087 NM 002673.3 215807 s at 17.47030156 1132835 SOXi1 30581115 NM 003108.3 204915 s at 16.89404131 1109505 MGC39372 19263721 BC025340.1 239186 at_ 15.78111902 1139054 ZBED5 49574209 NM 021211.2 218263 s at 15.77800815 1119361 TIAl 11863160 NM 022037.1 201448 at 15.68070962 1115226 KIAA1683 144922706 NM 025249.2 223600 s at 15.67954057 1101211 STRBP 21361744 NM 018387.2 229513 at 15.4183527 1118963 11084506 BF196503.1 65472 at 15.36802586 1096503 C9orf45 12005727 AF251293.1 223522 at 14.64776335 1127849 SNN 83627728 NM 003498.4 218032 at 14.54859775 1099204 5817123 AL110204.1 227121 at 14.32724822 1098840 CCDC50 41281910 NM 178335.1 226713 at 14.10346944 1139444 RABL2B 51317348 NM 001003789.1 219151 s at 14.10016196 1106855 KIAA1909 148529024 NM 052909.3 236255 at 13.9504946 1126695 LOC730273 113419660 XM 001124203.1 216000 at 13.92285415 1120137 FCGBP 154146261 NM 003890.2 203240 at 13.86147896 1133011 TMSL8 72255577 NM 021992.2 205347 s at 13.74377784 1133192 RASGRP3 24762238 NM 170672.1 205801 s at_ 17.09085725_ Table 27: MCL vs. BL predictor genes Gene GenBank Affymetrix UNIQID symbol gi .Accession _probe set ID Scale Factor 1120900 EPHB6 56119211 NM 004445.2 204718 at 13.43582327 116 Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1112061 SDK2 48255893 NM 019064.3 242064 at 12.73065392 1109505 MGC39372 19263721 BC025340.1 239186 at 12.63674985 1133099 DNASEL3 58331226 NM 004944.2 205554 s at 12.43333984 1106855 KIAA1909 148529024 NM 052909.3 236255 at 12.32623489 1110070 8168250 AW977010.1 239803 at 12.05416064 1121739 ZNF135 34419632 NM 003436.2 206142 at 11.90460363 1098840 CCDC50 41281910 NM 178335.1 226713 at 11.90309143 1132833 SOXI1 30581115 NM 003108.3 204913 s at 11.60864812 1121693 PLCH2 78499632 NM 014638.2 206080 at 11.33634052 1123760 LILRA4 47519952 NM 012276.3 210313 at 11.18744726 1125964 TMEM63A 7662307 NM 014698.1 214833 at 11.14762675 1112306 6139540 AW135407.1 242354 at 11.02434114 1096070 DNMT3A 28559066 NM 022552.3 222640 at 10.98991879 1129943 ZNF506 149944539 NM 001099269.1 221626 at 10.72494956 1118749 11125347 AJ296290.1 39313 at 10.64623382 1098954 MOBKL2B 41350329 NM 024761.3 226844 at 10.46164401 1134749 ZMYND8 34335265 NM 012408.3 209049 s at 10.40948157 1131860 BINI 21536381 NM 004305.2 202931 x at 10.31084561 1123148 TGFBR2 133908632 NM 001024847.2 208944 at 10.2956213 Table 28: MCL vs. FH predictor enes Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1132834 SOXi1 30581115 NM 003108.3 204914 s at 24.3531072 1100873 WNT3 21536426 NM 030753.3 229103 at 16.83342764 1109603 4971784 A1694444.1 239292 at 13.02401995 1139411 OSBPLIO 23111057 NM 017784.3 219073 s at 12.54369577 1106855 K1AA1909 148529024 NM 052909.3 236255 at 12.10316361 1125193 CNR1 38683843 NM 016083.3 213436 at 12.070579 1137450 ALOX5 62912458 NM 000698.2 214366 s at 11.74571823 117 Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1100258 KLHL14 55741642 NM 020805.1 228377 at 11.60998697 1133167 ZNF107 62243639 NM 001013746.1 205739 x at 11.52931491 1136831 PPFIBP2 57163846 NM 003621.1 212841 s at 11.50062692 1138222 CD24 73623396 NM 013230.2 216379 x at 10.99674674 1099437 PTPRJ 148728159 NM 001098503.1 227396 at 10.90797288 1140236 FCRL2 74048626 NM 138738.2 221239 s at 10.77082801 1114109 CLECLI 40548404 NM 172004.2 244413 at 10.65867119 1098277 PRICKLEI 23308518 NM 153026.1 226065 at 10.55457068 1135138 CD24 73623396 NM 013230.2 209771 x at 10.41999962 1103304 LOC439949 113421296 XM 001128367.1 232001 at -10.46625233 1128460 PITPNCI 32307139 NM 012417.2 219155 at -10.91106245 1121953 KIAAO125 20302136 NM 014792.2 206478 at -11.22466255 1129281 FAM30A 6841343 AF161538.1 220377 at -15.54465448 Table 29: MCL vs. FL predictor genes Gene GenfBank Affymetrix Scale UNIQID symbol gi Accession probe set ID Factor 1132835 SOXI1 30581115 NM 003108.3 204915 s at 22.14208817 1096070 DNMT3A 28559066 NM 022552.3 222640 at 20.53740132 1103711 10433184 AK021895.1 232478 at 20.49880004 1137987 PLXNBI 41152087 NM 002673.3' 215807 s at 18.38081568 1109505 MGC39372 19263721 BC025340.1 239186 at 17.17812448 1098840 CCDC50 41281910 NM 178335.1 226713 at 16.32703666 1130926 C5orfl3 4758865 NM 004772.1 201310 s at 15.34261878 1096396 SPG3A 74024913 NM 015915.3 223340 at 14.75437736 1132734 COL9A3 119508425 NM 001853.3 204724 s at 14.684583 1139393 OPN3 71999130 NM 014322.2 219032 x at 14.39118445 1115537 CNFN 26024194 NM 032488.2 224329 s at 14.18446144 1102215 PRICKLEl 23308518 NM 003105.3 230708 at 14.16246426 118 Gene GenBank Affymetrix Scale UNIQID symbol gi Accession probe set ID __Factor 1124585 SORLI 18379347 NM 003105.3 212560 at 14.33315955 1137561 HOXAl 84697023 NM 005522.4 214639 s at 15.38404642 1100581 BCL6 21040323 NM 001706.2 228758 at 15.91666634 1124646 RFTN1 41872576 NM 015150.1 212646 at 16.40577696 1114543 5395779 A1809213.1 244887 at 17.60167863 1120090 BCL6 21040323 NM 001706.2 203140 at 17.63091181 1123731 RGS13 21464137 NM 170672.1 210258 at 22.41602151 1133192 RASGRP3 24762238 NM 170672.1 205801 s at 27.28308723 Table 30: MCL vs. GB predict genes _______.. Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1098840 CCDC50 41281910 NM 178335.1 226713 at 22.26488562 1132835 SOXi1 30581115 NM 003108.3 204915 s at 17.76179754 1137987 PLXNB1 41152087 NM 002673.3 215807 s at 16.86845147 1098954 MOBKL2B 41350329 NM 024761.3 226844 at 16.65023669 1103711 10433184 AK021895.1 232478 at 15.64719784 1096070 DNMT3A 28559066 NM 022552.3 222640 at 15.22540494 1139393 OPN3 71999130 NM 014322.2 219032 x at 14.64030565 1127849 SNN 83627728 NM 003498.4 218032 at 14.28242206 1098156 MAP3K1 153945764 NM 005921.1 225927 at 14.00049272 1128845 SIDTI 116812583 NM 017699.2 219734 at 13.96064416 119 Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1129943 ZNF506 149944539 NM 001099269.1 221626 at 13.85404507 1140116 ARHGAP24 111154091 NM 001025616.2 221030 s at 13.81464172 1106855 KIAA1909 148529024 NM 052909.3 236255 at 13.74521849 1120900 EPHB6 56119211 NM 004445.2 204718 at 13.46567004 1127371 4372199 A1479031.1 217164 at 13.45735668 1119361 TIAl 11863160 NM 022037.1 201448 at 13.37376559 1120854 EDGI 87196352 NM 001400.3 204642 at 13.1047657 1098277 PRICKLEl 23308518 NM 153026.1 226065 at 13.04993076 TRIM6 1140127 TRIM34 51477689 NM 001003819.1 221044 s at 12.66260609 1100581 BCL6 21040323 NM 001706.2 228758 at -12.81251689 Table 31: MCL vs. ALT predictor genes Gene GenBank Affymetrix Scale UNIQID symbol gi Accession probe set ID Factor 1132834 SOXI1 30581115 NM 003108.3 204914 s at 20.7489202 1101987 KIAA1909 148529024 NM 052909.3 230441 at 10.78991326 1100873 WNT3 21536426 NM 030753.3 229103 at 10.11845036 1130764 HNRNPAO 52426775 NM 006805.3 201055 s at 9.432459453 1102178 H2BFXP 115432116 NR 003238.1 230664 at 9.035605572 1098277 PRICKLEI 23308518 NM 153026.1 226065 at 9.003360784 1130926 C5orfI3 4758865 NM 004772.1 201310 s at 8.712830747 1098694 SBK1 67906173 NM 001024401.2 226548 at 8.309789856 1103711 10433184 AK021895.1 232478 at 8.248526605 1138099 FADS3 34304362 NM 021727.3 216080 s at 8.107440225 1120854 EDGI 87196352 NM 001400.3 204642 at 8.045872672 1102215 PRICKLEl 23308518 NM 153026.1 230708 at 8.032351578 1121739 ZNF135 34419632 NM 003436.2 206142 at 8.020919565 1096070 DNMT3A 28559066 NM 022552.3 222640 at 7.964477216 1101211 STRBP 21361744 NM 018387.2 229513 at 7.738742472 120 Gene GenBank Affymetrix Scale UNIQID symbol gi Accession probe set ID Factor 1120825 CHLI 27894375 NM 006614.2 204591 at 7.516130116 1099437 PTPRJ 148728159 NM 001098503.1 227396 at 7.209041652 1096503 C9orf45 12005727 AF251293.1 223522 at 7.171540413 1135927 LILRA2 5803067 NM 006866.1 211102 s at 7.134470829 1120645 FADS3 34304362 NM 021727.3 204257 at 7.039952979 Table 32: MCL vs. PMBL predictor genes Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1132834 SOX11 30581115 NM 003108.3 204914 s at 28.17593839 1100873 WNT3 21536426 NM 030753.3 229103 at 17.90004832 1096503 C9orf45 12005727 AF251293.1 223522 at 17.43982729 1098840 CCDC50 41281910 NM 178335.1 226713 at 17.37421052 1124734 ZNF238 45439300 NM 006352.3 212774 at 16.73821457 1135102 PRKCB1 47157320 NM 002738.5 209685 s at 16.67436366 1103711 10433184 AK021895.1 232478 at 16.57202026 1140416 FAIM3 34147517 NM 005449.3 221601 s at 15.64802242 1121757 ADRB2 116686129 NM 000024.4 206170 at 15.57336633 1140236 FCRL2 74048626 NM 138738.2 221239 s at 15.20264513 1099140 28835721 BC047541.1 227052 at 15.11929571 1099549 7154210 AW516128.1 227533 at 14.92883027 1139054 ZBED5 49574209 NM 021211.2 218263 s at 14.63422275 1138818 ILF3 24234752 NM 004516.2 217804 s at 14.50621028 1109444 31317247 NM 006850.2 239122 at 14.20430672 GPATC NM_0010029 1124534 H8 50962881 09.1 212485 at 14.18537487 PRICKL 1098277 El 23308518 NM 153026.1 226065 at 13.98526258 1131687 TLK1 33636697 NM 012290.3 202606 s at 13.97468703 1125112 PLCL2 142369931 NM 015184.3 213309 at 13.85714318 1 20a Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1125397 RABL4 9257237 NM 006860.2 213784 at 13.85049805 Table 33: MCL vs. PTLD predictor genes Affymetrix Gene GenBank probe set Scale UNIQID symbol gi Accession ID Factor 1109603 14060141 BG749488.1 239292 at 19.95553782 1138222 CD24 73623396 NM 013230.2 216379 x at 15.95397369 1135138 CD24 73623396 NM 013230.2 209771 x at 15.89198725 1134230 RASGRP2 149158726 NM 001098670.1 208206 s at 15.80452978 1139411 OSBPL1O 23111057 NM 017784.3 219073 s at 14.32818885 1140416 FAIM3 34147517 NM 005449.3 221601 s at 13.89685188 1132834 SOXI1 30581115 NM 003108.3 204914 s at 13.78424818 1121739 ZNF135 34419632 NM 003436.2 206142 at 13.02195529 1098156 MAP3K1 153945764 NM 005921.1 225927 at 12.95032505 1099270 AFF3 68348715 NM 001025108.1 227198 at 12.7877735 1139012 MAP4K4 46249361 NM 004834.3 218181 s at 12.70176225 1120854 EDG1 87196352 NM 001400.3 204642 at 12.25264341 1120985 ARHGAP25 55770897 NM 001007231.1 204882 at 12.04626201 1115952 ATXNIL 21734020 AL833385.1 226095 s at 11.96299478 1120825 CHLI 27894375 NM 006614.2 204591 at 11.82402907 1131636 SPOCK2 7662035 NM 014767.1 202524 s at 11.80417657 1136706 PCMTD2 157388996 NM 001104925.1 212406 s at 11.74962191 1113560 14589868 NM 032966.1 243798 at 11.72049882 1133851 P4HA 63252885 NM 000917.2 207543 s at 12.59876059 1137459 BCAT1 72187658 NM 005504.4 214390 s at 14.00465411 12 0 b Table 34: MCL vs. SLL predictor genes Gene GenBank Affymetrix UNIQID symbol gi Accession probe set ID Scale Factor 1132834 SOXI1 30581115 NM 003108.3 204914 s at 23.59602107 1101987 KIAA1909 148529024 NM 052909.3 230441 at 14.50254794 1103711 10433184 AK021895.1 232478 at 13.31375894 1096070 DNMT3A 28559066 NM 022552.3 222640 at 12.37453972 1130926 C5orfI3 4758865 NM 004772.1 201310 s at 11.27840239 1120645 FADS3 34304362 NM 021727.3 204257 at 11.14057287 1138099 FADS3 34304362 NM 021727.3 216080 s at 10.92729287 1097887 MAST4 148727254 NM 015183.1 225611 at 10.37913127 1099941 15750678 B1759100.1 227985 at 10.33953409 1130373 MAST4 148727254 NM 015183.1 40016 g at 10.01524528 1110957 SYNE2 118918402 NM 015180.4 240777 at 9.865436185 1130320 MAST4 148727254 NM 015183.1 222348 at 9.807091644 1124373 LPINI 22027647 NM 145693.1 212274 at 9.024985551 1128813 KREMEN2 27437002 NM 024507.2 219692 at 8.903791941 1131130 MARCKS 153070259 NM 002356.5 201670 s at 8.688979176 1120825 CHL 27894375 NM 006614.2 204591 at 8.685132271 1119752 BASPI 30795230 NM 006317.3 202391 at 8.663402838 GCLC 45359851 NM_001498.2 202923_s_at 1131854 8.761521136 MAN2A1 51477713 NM_002372.2 235103_at 1105801 '' 8.828675125 MAP2 87578393 NM_001039538.1 225540_at 1097824 9.345688564 I 20c Table 35: MCL vs. s plenic predicto- genes Gene GenBank Affymetrix UJNIQID symbol Ai Accession probe set ID Scale Factor 1106855 KIAA1909 148529024 NM 052909.3 236255 at 14.48278638 1121739 ZNF135 34419632 NM 003436.2 206142 at 11.95918572 1111850 BC047698.1 BC047698.1 241808 at 11.13464157 1098024 RSPRYI 45387948 NM 133368.1 225774 at 10.10869886 1130764 HNRNPA0 52426775 NM 006805.3 201055 s at 10.06898534 1135342 SHOX2 87044887 NM 003030.3 210135 s at 9.565884385 1097218 TCEAL8 55749465 NM 001006684.1 224819 at 9.187725705 1117193 ZBTB1O 157694498 NM 001105539.1 233899 x at 9.12522795 1139564 PSMD1O 28605122 NM 002814.2 219485 s at 9.066714773 1132834 SOXI1 30581115 NM 003108.3 204914 s at 8.908574745 1131130 MARCKS 153070259 NM 002356.5 201670 s at 8.732921026 1131756 PDCD4 34304340 NM 014456.3 202730 s at 8.441424593 1102187 PKHDIL1 126116588 NM 177531.4 230673 at 8.391861029 1098195 TMEM64 116089278 NM 001008495.2 225974 at 8.349839204 1101211 STRBP 21361744 NM 018387.2 229513 at 8.337208237 1136673 GNAS 117938757 NM 000516.4 212273 x at 8_254076655_ 1139116 USP16 50312663 __M 00019 . 28386 at8.7345 1098694 SBKI 67906173 __NM_0010_24401.2 226548_at .9_038 1120519 WWP2 40806206 NM 007014.3 204022 at 7.881202253 1114916 CPLX3 , 75677370 NM 001030005.2 222927 s at -8.336831 9 12 1 With so many candidate predictor genes being utilized, it is possible to generate a predictor model that accurately predicts every element of a training set but fails to perform on an independent sample. This occurs because the model incorporates and "learns" the individual characteristics of each sample in the training 5 set. Leave-one-out cross-validation was used to verify that the prediction models generated above would work on independent samples that the models had not encountered previously. In this cross-validation method, a single sample is removed from the training set, and the predictor is developed again using the remaining data. The resulting model is then used to predict the sample that was removed. This 10 method is repeated with each individual sample taken out. Since no sample is predicted from a model that includes that sample, this method provides an unbiased estimate of predictor accuracy. When the predictors developed above were evaluated by leave-one-out cross-validation, all but one of the 21 MCL samples were correctly identified as MCL ,5 and none of the 489 non-MCL samples were mistakenly identified as MCL. Example 12: Identification of lymphoma samples as MCL based on Bayesian analysis of gene expression data from a Lyriphochip microarray: Lymphoma samples with morphology consistent with MCL were identified by pathological review. Since t(11 ;14) translocation and cyclin D1 overexpression have 20 been consistently associated with MCL, cyclin Di mRNA levels were measured in each sample by quantitative RT-PCR. Of the 101 samples analyzed, 92 expressed cyclin D1 mRNA. These 92 samples, which were deemed the "core group" of MCLs, were divided into a training set and a validation set. Gene expression was measured in all 101 samples using a Lymphochip microarray (Alizadeh 1999). For 122 comparison, gene expression was measured in 20 samples identified as SLL. In addition, MCL expression data was compared to expression data obtained previously for GCB (134 cases) and ABC (83 cases) (Rosenwald 2002). Several thousand genes were differentially expressed between cyclin D1-positive MCL and 5 the other lymphoma types with high statistical significance (p < 0.001). A complete listing of these genes is available at http://llmpp.nih.gov/MCL. Three different binary predictor models were developed: MCL vs. SLL, MCL vs. GCB, and MCL vs. ABC. Each of these models was designed to calculate the probability that a sample was MCL rather than the other lymphoma type in the pair. 10 For each pair, the genes that were most differentially expressed between MCL and the other lymphoma type in the pair were identified, and the difference in expression between the lymphoma types was quantified using a Student's t-test. An LPS was then calculated for each sample using the following formula: LPS(X) = t X , leG .5 where X is the expression of gene j in sample X and t; is the t-statistic for the difference in expression of gene j between the two lymphoma types in the pair. Cyclin D1 was excluded from the calculation of LPS so that the model could be used to identify potential MCL cases that were cyclin D1 negative. After an LPS had been formulated for each lymphoma sample, the mean and 20 standard deviation of these LPS's was calculated for each lymphoma type. For a new sample X, Bayes' rule can be used to estimate the probability q that the sample belongs to MCL rather than the second lymphoma type in the pair using the following equation: 123 q(X is MCL) #(LPS(X); + bMCS ) #(LPS(X); AA CL * MCL )+#(LPS(X); A2, d2) where #(x; p,o-) is the normal density function with mean p and standard deviation a, AMCL ad &MCL are the sample mean and variance of the LPS values for MCL, and A 2 andd- 2 are the sample mean and variance of the LPS values for the 5 second lymphoma type of the pair. A cut-off point of 90% was selected for assigning a sample to a particular lymphoma type. Every sample in the training set were classified correctly using this model (Figure 16). When applied to the validation set, the model correctly classified 98% of the cyclin D1-positive MCL cases as MCL (Figure 16). t0 This diagnostic test was applied to nine lymphoma cases that were morphologically consistent with MCL, but negative for cyclin D1 expression. Seven of these samples were classified as MCL, one was classified as GCB, and one was not assigned to any lymphoma type because none of the pairs generated a probability of 90% or greater. i5 Example 13: Classification of DLBCL samples based on Bayesian analysis of gene expression data from the Lymphochip microarray: A statistical method to classify DLBCL samples based on Bayesian analysis was developed using gene expression data obtained using the Lymphochip cDNA microarray (Rosenwald 2002). This data is available at http://llmpp.nih.gov/DLBCL. 20 The data was divided into two sets: a training set used to create and optimize the prediction model, and a validation set to evaluate the performance of the model. The training set consisted of 42 ABC DLBCL samples and 67 GCB DLBCL samples, 124 while the validation set consisted of 41 ABC DLBCL samples, 67 GCB DLBCL samples, and 57 type 3 DLBCL samples (Shipp 2002). Genes that were listed as present on >50% of the samples were identified, and the signal value for these genes on each microarray was normalized to 1,000. 5 After normalization, all signal values under 50 were set to 50. A log2 transformation was then performed on all the signal values. An LPS for distinguishing between two lymphoma types was calculated for each sample X in the training set using an equation: LPS(X) = tjXj, 10 where X represents the expression level of gene j and tj is a scaling factor whose value depends on the difference in expression of gene j between the two lymphoma types. The scaling factor used in this example was the t-statistic generated by a t test of the difference in gene j expression between two lymphoma types. Only those genes with the largest t-statistics were included when calculating the LPS for each 15 sample. The list of genes used to generate the LPS was narrowed further by including only those genes that were most variably expressed within the training set. Only genes in the top third with respect to variance were included. Genes that displayed a correlation with proliferation or lymph node signatures (Shaffer 2001; Rosenwald 2002) were eliminated from consideration, because these genes are 20 often variably expressed within samples from a single lymphoma type (Rosenwald 2002). Since the LF$S is a linear combination of gene expression values, its distribution within each lymphoma type should be approximately normal, provided that it includes a sufficient number of genes and the correlation structure of those 125 genes is not extreme. The mean and variance of these normal distributions within a lymphoma type can then be estimated from the combined LPS's of all samples within the type. The LPS distribution of two lymphoma types can be used to estimate the probability that a new sample belongs to one of the types using Bayes' rule. The probability q that a s sample Y belongs to lymphoma type I can be determined by an equation: q(Y is subtype 1) =(LPS(Y); A #(LPS(Y); A,,6,)+#(LPS(Y); p2,0-2) where $(x; p,o-) is the normal density function with mean p and standard deviation a, A, and 6, are the sample mean and variance of the LPS values for lymphoma 10 type 1, and$ 2 and &2 are the sample mean and variance of the LPS values for lymphoma type 2. This calculation was used to determine the probability that each sample in the training set belonged to GCB or ABC. A sample was classified as a particular type if it had a 90% or greater probability of belonging to that type. The number of genes in the predictor model was optimized based on the accuracy with which the predictor is classified samples into the ABC or GCB subtypes defined previously by hierarchical clustering (Rosenwald 2002). The final predictor incorporated 27 genes, and correctly classified 87% of the training set samples into the subtype to which they had been assigned by hierarchical clustering (Figure 17). The genes included in the predictor are listed in Table 36. Table 36 Unigene ID Build 167 (http://www.ncbi.nlm.nih.gov Gene GenBank Affymetrix UNIQID /UniGene) symbol gi Accession probe set ID 19375 235860 FOXPI 60498986 NM 001012505.1 224837 at 19346 109150 SH3BP5 109134343 NM 001018009.2 201810 s at 19227 193857 LOC96597 89041248 XM 378655.2 233483 at 16049 439852 IGHM 21757751 AK097859.1 32529 55098 CCDC50 33186926 NM 174908.2 226713 at 24729 127686 IRF4 4505286 NM 002460.1 204562 at 24899 81170 PIMI1 31543400 NM 002648.2 209193 at 126 19348 NA IGHM 21757751 AK097859.1 27565 444105 ENTPD1 147905699 NM 001098175.1 209474 s at 17227 170359 IL16 148833502 NM 004513.4 209827 s at 26919 118722 FUT8 30410721 NM 004480.3 203988 s at 24321 171262 ETV6 153267458 NM 001987.4 217377 x at 29385 167746 BLNK 40353774 NM 013314.2 207655 s at 16858 376071 CCND2 16950656 NM 001759.2 200951 s at 31801 386140 BMF 51558687 NM 001003940.1 226530 at 19234 418004 PTPN1 18104977 NM 002827.2 202716 at 26385 307734 . MME 116256328 NM 000902.3 203434 s at 24361 388737 76779240 BC106050.1 24570 446198 SAMD12 156119596 NM 001101676.1 24904 18166 KIAA0870 50345869 NM 014957.2 212975 at 24429 155024 BCL6 21040323 NM 001706.2 228758 at 28224 387222 NEK6 34147501 NM 014397.3 223158 s at 27673 124922 LRMP 42789728 NM 006152.2 204674 at 24376 317970 SERPINA11 110225348 NM 001042518.1 17496 300592 MYBLI 122937230 NM 001080416.1 213906 at 17218 283063 LMO2 6633806 NM 005574.2 204249 s at 28338 78877 ITPKB 38569399 NM 002221.2 203723 at Since the samples used to estimate the distribution of the LPS's were the same samples 15 used to generate the model, there was a possibility of overfitting. Overfitting would result in a model that indicates a larger separation between the LPS's of two lymphoma types than would be found in independent data. To ensure that overfitting was not taking place, the model was tested on the validation set. The reproducibility of the predictor model was verified by its ability to correctly classify 88% of the samples in the validation set (Figure 20 18). Interestingly, 56% of the DLBCL samples that had been placed in the type 3 subtype by hierarchical clustering were classified as either ABC or GCB using this Bayesian model. In previous experiments, the genes that were used to distinguish GCB and ABC were deliberately selected to include those that were preferentially expressed in normal 25 GC B cells (Alizadeh 2000; Rosenwald 2002). In the present analysis, the predictor model was not biased a priori to include such genes. The ABC and GCB lymphoma types as defined by the Bayesian model were analyzed for differential 127 expression of GC B cell restricted genes. Thirty seven genes were found to be both more highly expressed in GC B cells than at other stages of differentiation (p<0.001) and differentially expressed between DLBCL subtypes (p<0.001) (Figure 19A). These 37 genes are listed in Table 37. 5 Table 37 UNIQID Unigene ID Build 167 Gene symbol (http://www.ncbi.nlm.nih.gov /UniGene) 28014 300592 MYBL1 24376 317970 SERPINA11 24429 155024 BCL6 16886 124922 LRMP 27374 283063 LMO2 29912 446198 24510 266175 PAG 24854 439767 TOX 32171 307734 MME 24361 388737 19365 171857 Cyorf15a 27292 272251 KLHL5 24822 283794 PCDHGC3 30923 446195 24825 88556 HDAC1 31696 91139 SLCIA1 26976 434281 PTK2 19279 49614 GCET2 17866 1765 LCK 24386 437459 MYO1E 33013 293130 VNN2 25126 30498 157441 Spi1 26512 379414 MFHAS1 26582 153260 SH3KBP1 17840 132311 MAP2K1 26000 25155 NETI 24323 149342 AICDA 30922 435904 C21lorflO 7 30641 79299 LHFPL2 19308 179608 DHRS9 24455 405387 30034 300208 SEC231P 24977 169939 HS2ST1 24449 206097 RRAS2 30763 446198 27987 73792 CR2 All but two (AICDA and DHRS9) of these 37 genes were more highly expressed in GCB than in ABC. This demonstrates that the DLBCL subtypes defined by the 128 Bayesian predictor seem to differ with respect to their cell of origin, with GCB retaining the gene expression program of normal GC B cells. ABC, on the other hand, displayed higher expression of genes characteristic of plasma cells (Figure 19B). Twenty four genes were found to be both more highly 5 expressed in plasma cells than in B cells at earlier developmental stages (p<0.001) and differentially expressed between the DLBCL subtypes (p<0.001). These 24 genes are listed in Table 38. Table 38 UNIQID Unigene ID Build 167 Gene symbol (http://www.ncbi.nlm.nih.gov /UniGene) 16614 127686 - IRF4 26907 118722 FUT8 31104 313544 NS 19219 355724 CFLAR 26174 28707 SSR3 24566 169948 KCNA3 34500 442808 B4GALT2 26991 314828 UPPi 30191 438695 FKBP11 27402 259855 EEF2K 26096 434937 PPIB 15887 2128 DUSP5 32440 512686 C20orf59 34827 429975 PM5 29232 437638 XBPI 17763 76640 RGC32 32163 445862 RAB30 17814 5353 CASPIO 31460 409223 SSR4 26693 83919 GCS1 25130 409563 PACAP 16436 267819 PPP1R2 31610 76901 PDIR 28961 212296 ITGA6 10 The majority of these plasma cell-restricted genes were more highly expressed in ABC than in GCB. Eight of the 32 genes encode proteins that reside and function in the endoplasmic reticulum (ER) or Golgi apparatus, suggesting that ABCs have increased the intracellular machinery for protein secretion. These eight genes are 129 denoted in the above list by the designation "ER" or "golgi" in parentheses. Another gene on this list, XBP- 1 transcription factor, encodes a protein that is required for plasma cell differentiation (Reimold 2001 ) and is involved in the response to unfolded proteins in the ER (Calfon 2002). ABC have not undergone full plasmacytic differentiation, however, 5 because other key plasma cell genes such as Blimp-I were not more highly expressed in ABC. Example 14: Classification of DLBCL samples based on Bayesian analysis of gene expression data from the Affymetrix HU6800 microarray: The prediction method described in Example 13 above was applied to gene to expression data from 58 DLBCL samples obtained using an Affymetrix HU 6800 oligonucleotide microarray (Shipp 2002). This data is available at www.genome.wi.mit.edu/MPR/lymphoma. The first step in analyzing this data was to exclude all microarray features with a median signal value of <200 across the samples. Multiple microarray features representing the same gene were then averaged. Of the 27 is genes in the DLBCL subtype predictor developed using the Lymphochip data (above), only 14 were represented on the Affymetrix array and passed this filtering process. These 14 genes are listed in Table 39. Table 39 Unigene ID Build 167 Gene GenBank Affymetrix UNIQID (http://www.ncbi.nm.nih.gov/UniGene) symbol gi Accession probe.set ID 24729 127686 IRF4 4505286 NM 002460.1 204562 at 17227 170359 1L16 148833502 NM 004513.4 209827 s at 26907 118722 FUT8 30410721 NM 004480.3 203988 s at 27565 444105 ENTPD1 147905699 NM 001098175.1 209474 s at 16858 376071 CCND2 16950656 NM 001759.2 200951 s at 24899 81170 PIMi 31543400 NM 002648.2 209193 at 16947 418004 PTPN1 18104977 NM 002827.2 202716 at 16049 439852 IGHM 21757751 AK097859.1 26385 307734 MME 116256328 NM 000902.3 203434 s at 27673 124922 LRMP 42789728 NM 006152.2 204674 at 24429 155024 BCL6 21040323 NM 001706.2 228758 at 17218 283063 LMO2 6633806 NM 005574.2 204249 s at 28338 78877 ITPKB 38569399 NM 002221.2 203723 at 17496 300592 MYBL1 122937230 NM 001080416.1 213906 at 130 These 14 genes were used to create a new DLBCL subtype predictor in which the LPS scaling coefficients were again calculated based on the DLBCL subtype distinction in the Lymphochip data set (Rosenwald 2002). To account for systematic measuring differences between the Affymetrix and Lymphochip microarrays, the expression value of each gene 5 on the Affymetrix microarray was shifted and scaled to match the mean and variance of the corresponding expression values on the Lymphochip. The adjusted expression values for each of the 14 genes were then used to calculate LPS's for each sample. DLBCL subtype membership was again assigned on a cut-off of 90% certainty. Several observations suggested that the predictor identified ABC and GCB samples within the io Affymetrix data set that were comparable to those found in the Lymphochip data set. First, the relative proportions of ABC (29%) and GCB (53%) were very similar to the corresponding proportions in the Lymphochip data set (34% and 49%, respectively). Second, 43 genes were found to be differentially expressed between the two DLBCL subtypes with high significance (p < 0.001) in the Affymetrix data. This number is is substantially higher than would be expected by chance, given that the Affymetrix microarray measures the expression of approximately 5,720 genes. The symbols for these 43 genes were: IGHM; TCF4; IRF4; CCND2; SLA; BATF; KLAAO171 ; PRKCB1; P2RX5; GOT2; SPIB; CSNK1E; PIM2; MARCKS; PIMI; TPM2; FUT8; CXCR4; SP140; BCL2; PTPNI; KIAA0084; HLA-DMB; ACP1; HLA-DQA1; RTVP1; VCL; 20 RPL21; ITPKB; SLAM; KRT8; DCK; PLEK; SCAI; PSIP2; FAM3C; GPR18; HMG14; CSTB; SPINK2; LRMP; MYBLI; and LM02. Third, the 43 genes differentially expressed between the types included 22 genes that were not used in the predictor but were 13 1 represented on Lymphochip arrays. Fourteen of these 22 genes were differentially expressed on the Lymphochip array with high statistical significance (p < 0.001). Finally, the expression of the c-rel gene was previously found to correspond to amplification of the c-rel genomic locus in DLBCL tumor cells, and oncogenic event 5 occurring in GCB but not ABC (Rosenwald 2002). In the Affymetrix data set, c-rel was differentially expressed between the two subtypes (p = 0.0025), and was highly expressed only in a subset of GCB's. Example 15: Identification of DLBCL samples as PMBL based on Bayesian analysis of gene expression data from the Lymphochip microarray: L0 310 lymphoma biopsy samples identified as DLBCL by a panel of hematopathologists were divided into a 36 sample training set and a 274 sample validation set, with the validation set consisting of the DLBCL samples classified previously in Example 13. All patients from whom the samples were derived had been treated with anthracycline-containing multiagent chemotherapy protocols, with 5 some patients additionally receiving radiation therapy. The training set was profiled for gene expression using Lymphochip microarrays comprising 15,133 cDNA elements as described previously (Alizadeh 2000). This data is available at http://llmpp.nih.gov/PMBL. The validation set had previously been profiled using Lymphochip microarrays comprising 12,196 cDNA elements (Rosenwald 2002). 0 This data is available at http:/Ilmpp.nih.gov/DLBCL. A hierarchical clustering algorithm (Eisen 1998) was used to organize the genes by their expression patterns across the 36 samples in the training set. A large group of genes that were more highly expressed in lymphomas with mediastinal involvement than in other DLBCLs was shown to be tightly clustered in the resulting 132 dendrogram (Figure 20A). This cluster of genes included two genes, MAL and FIG1, previously shown to be highly expressed in PMBL (Copie-Bergman 2002; Copie Bergman 2003). Several of the lymphomas with mediastinal involvement did not express this set of putative PMBL signature genes, and it was suspected that these 5 samples were more likely to be conventional DLBCL than PMBL. Hierarchical clustering was used to organize the samples according to their expression of the PMBL signature genes, resulting in two major clusters of cases (Figure 20B). One cluster contained 21 samples designated "PMBL core" samples by virtue of their higher expression of PMBL signature genes. The other cluster contained some 0 samples that had virtually no expression of these genes, and other samples that did express these genes but at lower levels than the PMBL core samples. A gene. expression-based method for distinguishing PMBL core cases from GCB and ABC DLBCL cases based on Bayesian analysis was developed using the methods described in Examples 13 and 14. A set of genes were selected that were 5 differentially expressed between the PMBL core samples and both GCB and ABC (p < 0.001). This set of genes included all of the PMBL signature genes identified by hierarchical clustering (Figure 20A), as well as a large number of additional genes. Many of the genes in this set belonged to the lymph node gene expression signature (Alizadeh 2000; Rosenwald 2002). These genes were excluded from the final 0 predictor because they might cause some DLBCL samples with higher expression of lymph node gene expression signature genes to be misclassified as PMBL. The list of PMBL distinction genes was refined by adding a requirement that they also be differentially expressed between the PMBL core samples and a subgroup of six DLBCL samples with higher expression of lymph node gene expression signature 5 genes (p < 0.001). The resulting set of 46 genes included 35 genes that were more 133 highly expressed in PMBL and 11 genes that were more highly expressed in DLBCL (Figure 21A). The 46 genes in this set were PDL2, SNFT, IL13RA1, FGFR1, FLJ10420, CCL17/TARC, TNFRSF8/CD30, E2F2, MAL, TNFSF4/OX40 ligand, IL41 1/Fig1, IMAGE:686580, BST2, FLJ31131, FCER2/CD23, SAMSNI, JAK2, 5 FLJO0066, MSTIR, TRAFI, SLAM, LY75, TNFRSF6/Fas, FNBP1, TLR7, TNFRSF17/BCMA, CDKN1Alp21CIP1, RGS9, IMAGE:1340506, NFKB2, KIAA0339, ITGAM, 1L23A, SPINT2, MEF2A, PFDN5, ZNF141, IMAGE:4154313, IMAGE:825382, DLEU1, ITGAE, SH3BP5, BANK, TCL1A, PRKAR1B, and CARD11. A series of linear predictor scores were generated based on the expression of this 0 gene set. Based on the distribution of linear predictor scores within a particular lymphoma type, Bayes' rule can be used to estimate the probability that a particular sample belongs to either of the two types. An arbitrary probability cut-off of 90% or greater was used to classify a sample as a particular lymphoma type. All of the PMBL core samples were classified as PMBL using this method, as Were six of the 5 other lymphoma samples with mediastinal involvement. However, nine of the lymphoma samples with mediastinal involvement were classified as a DLBCL, as were all of the GCB and ABC samples. In the validation set, 11 samples were identified on clinical grounds as being consistent with a diagnosis of PMBL, and the Bayesian model classified nine of 0 these as PMBL (Figure 21B). Interestingly, 12 of the remaining 263 DLBCL samples were classified as PMBL by the predictor. Figure 21 B shows that these cases were indistinguishable by gene expression from the nine cases diagnosed as PMBL on clinical grounds. As expected, the average expression of the PMBL predictor genes in the 249 samples classified as DLBCL was notably lower than in the 22 PMBL 134 cases. Thus, PMBL represents a third subgroup of DLBCL than can be distinguished from ABC and GCB by gene expression profiling. Table 40 compares the clinical parameters of patients assigned to the PMBL, ABC, and GCB subgroups of DLBCL using this prediction method. 5 Table 40 ABC GCB PMBL PMBL PMBL P value DLBCL DLBCL Training set Validation set All cases Median age 66 61 33 33 33 4.4E-16 Age <35 5% 10% 52% 56% 53% 7.2E-14 Age 35-60 29% 38% 44% 28% 37% Age >60 66% 52% 4% 17% 9% Gender = male 59% 53% 44% 50% 47% 0.38 Female <35 2% 3% 32% 39% 35% 1.1 E-12 Male <35 2% 7% 20% 17% 19% Female 35-60 6% 18% 24% 6% 16% Male 35-60 23% 19% 20% 22% 21% Female >60 33% 25% 0% 6% 2% Male >60 34% 27% 4% 11% 7% PMBL patients were significantly younger than other DLBCL patients, with a median age at diagnosis of 33 years compared with a median age of 66 and 61 years for ABC and GCB patients, respectively. Although there was no significant difference in 0 gender distribution among the DLBCL subgroups, young women (< 35 years) accounted for 35% of PMBL patients, more than any other DLBCL subgroup. Young men (< 35 years) were also more frequently represented in the PMBL subgroup, accounting for 19% of the patients. Correspondingly, older men and women (age >60) were significantly underrepresented in the PMBL subgroup. These clinical 5 characteristics were observed in both the training set and the validation set of PMBL cases, demonstrating that the PMBL predictor reproducibly identified a clinically distinct subgroup of DLBCL patients. The PMBL subgroup defined by the PMBL predictor had a relatively favorable overall survival rate after therapy (Figure 22). PMBL patients had a five-year survival 135 rate of 64%, superior to the 46% rate seen in DLBCL patients as a whole (p = 0.0067). The survival of the PMBL subgroup was significantly better than the 30% five-year survival rate of the ABC subgroup (Figure 22; p = 5.8E-5), but only marginally better than the 59% five-year survival rate of the GCB subgroup (p = 5 0.18). Example 16: Classification of lymphomas into types based on Bayesian analysis of gene expression data from the Lymph Dx microarray: Based on the clustering of the Lymph Dx microarray signals for the DLBCL samples, a cluster of "proliferation signature" genes and a cluster of "lymph node 0 signature" genes were identified. The expression of these genes was averaged to form a proliferation signature and a lymph node signature. Each gene represented on the Lymph Dx microarray was placed into one of three "gene-list categories" based on its correlation with the proliferation or lymph node gene signatures. "Proliferation" genes were defined as those genes for which the correlation between 5 their expression and the proliferation signature was greater than 0.35. Lymph node genes were defined as those genes for which the correlation between their expression and the lymph node signature was greater than 0.35. The remaining genes on the array were classified as standard genes. This classification resulted in 323 proliferation genes and 375 lymph node genes. 0 Two stages of lymphoma classification were performed using the gene expression data obtained for the above samples using the Lymph Dx microarray. The general procedure used to classify the samples is presented in flow chart form in Figure 1.

136 For the first stage of expression analysis, the samples were divided into five types: FL, MCL, SLL, FH, and a class of aggressive lymphomas that included DLBCL and BL. Samples obtained from subjects with other diagnoses (e.g., MALT, LPC) were omitted from this analysis. Data from the Lymph Dx microarray was then 5 used to compare gene expression in each possible lymphoma type pair (e.g., FH vs. FL, MCL vs. SLL, etc.). This resulted in the creation of ten "pair-wise models" (one for each possible lymphoma type pair) for predicting whether a sample fell into a particular lymphoma type. For each lymphoma type pair, the difference in expression between the two 0 types for every gene on the microarray was calculated, and a t-statistic was generated to represent this difference. Within each gene-list category (proliferation, lymph node, and standard), individual genes were ordered based on the absolute value of their t-statistic. Only those genes that displayed a statistically significant difference in expression between the two types were included in the model. Those 5 genes with largest absolute t-statistics in each gene-list category were then used to generate a linear predictor score (LPS) for each sample. For a sample X and a set of genes G, the LPS was defined as: LPS(X) = t Xi, jeG where X is the expression of gene j in the sample and t is the t-statistic representing 0 the difference in expression of gene j between the two lymphoma types. This formulation of LPS, known as the compound covariate predictor, has previously been used successfully (Radmacher 2002; Rosenwald 2003; Wright 2003). Other ways to formulate an LPS include Fisher linear discriminant analysis (Dudoit 2002), 137 weighted voting (Golub 1999), linear support vector machines (Ramaswamy 2001), and nearest shrunken centroids (Tibshirani 2002). In order to optimize the number of genes used to generate the LPS, a series of LPS's were generated for each sample using between five and 100 genes from 5 each gene-list category. The optimal number of genes is that number which generates a maximum t-statistic when comparing the LPS of two samples from different lymphoma types (Figure 23). This optimization procedure was repeated for every gene-list category in every pair-wise model, meaning that 30 optimizations were performed in all. 0 It was recognized that for some pair-wise models, it would be useful to calculate LPS's using different combinations of gene-list categories. LPS's were calculated for each sample using four different combinations. In the first, LPS was calculated using the standard genes only. In the second, LPS's were calculated for both the standard and proliferation genes, but not the lymph node genes. In the 5 third, LPSs were calculated for both the standard and lymph node genes, but not the proliferation genes. In the fourth, LPS's were calculated using all three gene-list categories. Depending on the number of gene-list categories included, between one and three LPS's were calculated for each sample in the pair-wise models. Thus, each 20 sample could be thought of as a vector in a space of between one and three dimensions. Since the LPS's were sums of individual expressions, it was reasonable to approximate the distributions as normal. Multivariate normal distributions are defined by two quantities: a mean vector, which indicates the average value of each of the models within a given lymphoma type, and a 5 . covariance matrix, which indicates the magnitude and orientation spread of points 138 away from this center. Both of these quantities can be estimated empirically from the observed data. Figure 24 shows the Standard and Proliferation LPS's for the FL vs. DLBCL/BL pair-wise model. The dotted lines indicate the standard deviations from the fitted multivariate normal distributions. 5 Once the multidimensional distributions have been estimated, Bayes' rule (Bayes 1763) can be used to estimate the probability that a given sample belongs to one lymphoma type or another. Bayesian analysis of an LPS has been successfully employed in the past to distinguish DLBCL subtypes (Rosenwald 2003, Wright 2003). For a sample X, the probability q of the sample belonging to a first lymphoma 0 type rather than a second lymphoma type can be calculated using the formula: q q$(LPS(X); A,6i) #(LPS(X); A, 1 ) +#(LPS(X); 2,62) where LPS(X) is the linear predictor score for sample X, #(x; p, -) is the normal density function with mean p and standard deviation a-, A and 6d are the mean and 5 variance of the LPS's for the first lymphoma type, and /2 andd 2 are the mean and variance of the LPS's for the second lymphoma type. Using this equation, a single probability q value can be developed for each sample and for each of the four LPS combinations. This q value can then be used to classify a sample as a first lymphoma type, a second lymphoma type, or unclassified. Samples with the highest 0 q values are classified as the first lymphoma type, while samples with the lowest q values are classified as the second lymphoma type. Samples with middle range q values are deemed unclassified. Classifying the samples in this manner requires two cut-off points: a lower cut-off point between the second lymphoma type and 139 unclassified, and an upper cut-off point between unclassified and the first lymphoma type. To develop these cut-off points, samples were ordered by their q values, and each possible cut-off point between adjacent samples was considered. To ensure that the cut-off points were reasonable, the lower cut-off point was restricted to 5 between 0.01 and 0.5 and the upper cut-off point was restricted to between 0.5 and 0.99. Every cut-off point and model combination was analyzed by the following equation: 3.99 * [(% of type 1 misidentified as type 2) + (% of type 2 misidentified as [0 type 1)] + [(% of type 1 unclassified) + (% of type 2 misidentified)]. Using this equation, the cut-off point would be adjusted to allow an additional error only if this adjustment resulted in four or more unclassified samples becoming correctly classified. The final model and cut-off point for a given pair-wise analysis was that which minimized this equation. The equation utilizes percentages rather 5 than the actual number of cases in order to account for the different number of samples in each class. All cut-off points between a given pair of adjacent q-values will produce the same division of data. Since cut-off point optimality is defined in terms of dividing the data into subtypes, all cut-off points between a pair of borderline cases will be 20 equally optimal. In choosing where to place the actual cut-off point values, values were chosen that would lead to a larger unclassified region. When the lower cut-off point was being defined, a value would be chosen that was 1/5 of the way from the smallest borderline 6ase to the largest. When the upper cut-off point was being defined, a value would.be chosen that was 4/5 of the way from the smallest t5 borderline case to the largest. Figure 25 illustrates the q-results of optimizing the 140 cut-point for fhe FL versus DLBCL/BL samples. The optimal lower cut-off point for these samples was found at q=0.49, while the optimal upper cut-off point was found at q=0.84. Figure 26 indicates how this choice of cut-off points divided the space of LPS's. 5 The above procedures resulted in a series of pair-wise models for comparing every lymphoma type to every other lymphoma type. If there are n types, then there will be n-1 pair-wise models for each type. Since there were five lymphoma types in the stage I analysis, each type was involved in 4 pair-wise models. For instance, there were four different pair-wise models for MCL: MCL vs. FH, MCL vs. FL, MCL 0 vs. SLL, and MCL vs. DLBCL/BL. For each sample tested, each pair-wise model will produce one of three possible results: 1) the sample belongs to the first lymphoma type of the pair-wise model, 2) the sample belongs to the second lymphoma type of the pair-wise model, or 3) the sample is unclassified. If each of the n-1 models agrees that the sample belongs to a particular lymphoma type, then the sample is 5 designated as belonging to that type. If the n-1 models do not all agree that the sample belongs to a particular lymphoma type, the sample is designated as unclassified. To ensure that the above methods did not result in overfitting (i.e., models that fit particular idiosyncrasies of the training set but fail when applied to ,0 independent data), the models were validated by leave-one-out cross-validation fashion (Hills 1966). Each sample was removed from the data one at a time, and a predictive model was developed as described above using the remaining data. This model was then used to predict the sample that was removed. Since the model being used to predict a given sample was generated from data that did not include 14 1 that sample, this method provided an unbiased estimate of the accuracy of the model. The results of the leave-one-out predictions are set forth in Tables 41 and 42, below. The rows in each table correspond to different sample groups, while the 5 columns indicate the prediction results. The standard to which the prediction results were compared in this stage was the diagnoses of a panel of eight expert hematopathologists who used histological morphology and immunohistochemistry to classify the samples. Table 41 provides classification results for the five lymphoma types tested (DLBCL/BL, FL, FH, MCL, SLL), while Table 42 provides more specific 10 results for classification of subtypes within these five lymphoma types. The results set forth in Table 41 are also summarized in Figure 27. Table 41 DLBCUBL 249 6 0 0 0 7 262 95%, 2% 3% FL 5 154 0 0 0 14 173 89% 8% 3% FH 0 0 17 0 0 0 17 100% 0% 0% MCL 0 0 0 22 0 0 22 100% 0% 0% SLL 0 0 0 0 14 0 14 100 0% 0%.--O 15 Table 42 . .0 0 (0 ABC 78 0 0 0 0 0 78 100% 0% 0% GCB 77 4 0 0 0 4 85 91% 5% 5% PMBL 33 0 0 0 0 0 33 100% 0% 0% Unclassified 27 1 0 0 14 2 30 90% 7% 3% 142 DLBCL DLBCL (not yet 14 0 0 0 0 1 15 93% 7% 0% subclassed) BL 20 1 0 0 0 0 21 95% 0% 5% FLgrade1 1 78 0 0 0 3 82 95% 4% 1% FL grade 2 2 58 0 0 0 3 63 92% 5% 3% FL grade 3A 2 18 0 0 0 8 28 64% 29% 7% Combined FL 5 154 0 0 0 14 173 89% 8% 3% grades 1, 2, 3A FL grade 3B 2 1 0 0 0 4 7 14% 57% 29% FL unknown grade 3 11 0 0 0 0 14 79% 0% 21% FH 0 0 17 0 0 0 17 100% 0% 0% MCL 0 0 0 22 0 0 22 100% 0% 0% SLL 0 0 0 0 14 0 14 100% 0% 0% As seen in Table 41, perfect prediction of SLL, MCL, and FH samples was obtained. The success rate for predicting FL and the aggressive lymphomas (DLBCL/BL) was also very good, with only 3% of the samples being classified 5 incorrectly. As seen in Table 42, perfect prediction was also obtained for ABC and PMBL samples within the DLBCL samples. Example 17: Classification of DLBCL/BL samples into subtypes based on Bayesian analysis of gene expression data from the Lymph Dx microarray: Samples identified as DLBCL/BL in Example 16 were subdivided into four 10 types: ABC, GCB, PMBL, and BL. These samples were then used to generate six pair-wise models using the same procedure described in Example 16. The results of the leave-one-out. predictions using these pair-wise models are set forth in Table 43, below. These results are also summarized in Figure 28. The rows in the table correspond to different sample groups, while the columns indicate the prediction 15 results. In this stage, the ability of the prediction method to identify BL was again measured against the diagnoses of hematopathologists. The ability of the prediction method to identify the various DLBCL subtypes, on the other hand, was measured 143 against previous studies in which this distinction between subtypes was based on gene expression data from a Lymphochip microarray (Alizadeh 2000, Rosenwald 2002, Rosenwald 2003, Wright 2003). Table 43 V0 ABC 76 0 0 2 78 97% 3% 0 o GCB 1 66 2 4 4 77 86% 9% 5% PMBL 0 2 27 0 4 33 82% 12% 6% Unclassified DLB3CL 5 9 1 1 11 27 NA 41%. 4% DLBCL (not yet 5 5 0 1 3 14 NA 21% 7% sub classed)I 1B 0 1 0 18 1 20 90% 5% 5% FL grade 1 0 1 0 . 0 0 1 FL grade 2 0 1 0 0 1 2 FLgrade3A 0 0 1 3 0 0 2 Combined FL grades 1, 2,0 0 1 F~rd20 1 0 0 1 2 3A FL grade 3B 0 1 0 0 1 2 FL unknown grade 0 1 0 1 1 3 5 As seen in Table 43, only I of the 20 BL lymphoma samples was classified incorrectly. The classification of DLBCL into subtypes was also quite effective. All previously identified ABC subtype samples were again assigned to the ABC subtype, while only 5% of the GCB samples and 6% of the PMBL samples were assigned to a 0 different subtype than they were assigned to previously. The above classification was implemented using S+ software and the S+ subtype predictor script contained in the file entitled "Subtype_Predictor.txt," located in the computer program listing appendix contained on CD number 22 of 22. This S+ script implements the lymphoma prediction algorithm. When this script is pasted 5 into an S+ script window and run in a working directory containing the data set files 144 discussed below, it will produce a text file entitled "PredictionResults.txt," which indicates the results of the predictive algorithm. The other files in the computer program listing appendix contain the required data sets, in their required format, for carrying out the lymphoma type identification described above. The file entitled 5 "GeneData.txt" contains the gene expression values for each sample analyzed. This file is included in the working directory when the S+ subtype predictor script is run. The file entitled "GenelD.txt" contains information about the genes in the. GeneData.txt file, and is also included in the working directory when the S+ subtype predictor script is run. This file indicates the UNIQID for each gene, as well as the 0 extent to which the gene is associated with the lymph node and proliferation signatures ("LN.cor" and "pro.cor," respectively). The file entitled "SamplelD.txt" contains information about the samples included in the "GeneData.txt" file, specifically the original classification of all the samples. This file is also included in the working directory when the S+ subtype predictor script is run. The file entitled 5 "PredictionResults.txt" is an example of the productive output of the prediction algorithm. After the above model was validated using leave-one-out cross-validation, the model was re-fit using all of the data to generate a final predictor that could be applied to a new set of data. Tables 44-59, below, indicate for each of the pair wise 0 models the list of genes used, the weight given to each of those genes, the signature with which each gene was associated, the mean values and covariance matrices associated with the subtypes being compared, and the q-value cut-points of the pair wise model. Table 44: ABC vs. BL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene http:#/www.ncbi.nlm. symbol 145 nih.gov/UniGene Standard -18.87 1101149 517226 229437 at BIC Standard -17.4 1121452 227817 205681 at BCL2A1 Standard -16.42 1123163 421342 208991 at STAT3 Standard -16.2 1121629 41691 205965 at BATF Standard -15 1134095 89555 208018 s at HCK Standard -14.75 1132636 306278 204490 s at CD44 Standard -14.33 1119939 170087 202820 at AHR Standard -14.25 1100138 278391 228234 at TIRP Standard -14.02 1128626 501452 219424 at EB13 Standard -13.89 1132883 432453 205027 s at MAP3K8 Standard -13.88 1134991 444105 209474 s at ENTPD1 Standard -13.37 1109913 355724 239629 at CFLAR Standard -13.25 1120389 75367 203761 at SLA Standard -12.99 1131497 114931 202295 s at CTSH Standard -12.71 1115071 390476 223218 s at MAIL Standard -12.46 1136329 132739 211675 s at HIC Standard -12.41 1128195 115325 218699 at RAB7LI Standard -12.37 1124381 440808 212288 at FNBP1 Standard -12.30 1100562 26608 228737 at C20orf 100 Standard -12.24 1101272 179089 229584 at DKFZp434 Standard -12.18 1128536 21126 219279 at DOCK10 Standard -11.64 1098271 300670 226056 at CDGAP Standard -11.4.1 1119566 433506 201954 at ARPC1B Standard -11.11 -1120651 80205 204269 at PIM2 Standard -10.89 1098952 62264 226841 at KIAA0937 Standard -10.80 1099939 488173 227983 at MGC7036 Standard -10.67 1134270 352119 208284 x at GGT1 Standard -10.44 1134145 4750 208091 s at DKFZP564 Standard -10.39 1123437 73090 209636 at NFKB2 Standard -10.17 1119884 418004 202716'at PTPM1 Standard -10.14 1129269 62919 220358 at SNFT Standard -10.13 1126293 504816 215346 at TNFRSFS Standard -10.12 1112344 163242 242406 at Standard -10.10 1135550 221811 210550 s at RASGRFI Standard -10.08 1135165 170359 209827 s at IL16 Standard -10.05 1120808 127686 204562 at IRF4 Standard -10.01 1122087 72927 206693 at IL7 Standard -9.97 1132004 415117 203217 s at SIAT9 Standard -9.88 1114824 193370 222762 x at LIMDI Standard -9.87 1132034 410455 203271 s at UNC119 Standard -9.87 1099680 210387 227677 at JAK3 Standard -9.86 1132830 31210 204908 s at BCL3 Standard -9.79 1099631 367639 227624 at FLJ20032 Standard -9.78 1120267 256278 203508 at TNFRSF1B Standard -9.77 1124187 378738 211986 at MGC5395 Standard -9.73 1108970 140489 238604 at Standard -9.71 1136216 512152 211528 x at HLA-G Standard -9.71 1120993 327 204912 at IL10RA Standard -9.68 1100847 97411 229070 at C6orfIO5 Standard -9.64 1123413 418291 209575 at ILIORB Standard -9.62 1115704 350268 224569 s at IRF2BP2 Standard -9.58 1108237 126232 237753 at Standard -9.55 1121695 511759 206082 at HCP5 Standard -9.48 1101905 170843 230345 at I Standard -9.42 1119243 440165 201171 at ATP6VOE 146 Standard -9.39 1140457 210546 221658 s at IL21R Standard -9.32 1098506 193400 226333 at IL6R Standard -9.31 1139805 414362 220230 s at CYB5R2 Standard -9.30 1139037 173380 218223 s at CKIP-1 Standard -9.28 1130533 76507 200706 s at LITAF Standard -9.15 1098678 386140 226530 at BMF Standard -9.04 1133210 434374 205842 s at JAK2 Standard 9.05 1116432 409362 229356 x at KIAA1259 Standard 9.17 1097281 7037 224892 at PLDN Standard 9.17 1140018 438482 220917 s at PWDMP Standard 9.30 1119997 367811 202951 at STK38 Standard 9.41 1119817 409194 202561 at TNKS Standard 9.55 1139842 133523 220367 s at SAP130 Standard 9.64 1132122 307734 203434 s at MME Standard 9.77 1119258 88556 201209 at HDACI Standard 9.80 1128248 234149 218802 at FLJ20647 Standard 10.38 1101211 287659 229513 at STRBP Standard 10.52 1123419 170195 209590 at BMP7 Standard 10.71 1133755 404501 207318 s at CDC2L5 Standard 10.80 1128192 102506 218696 at EIF2AK3 Standard 10.85 1124786 22370 212847 at NEXN Standard 10.92 1130114 445084 221965 at MPHOSPH9 Standard 11.00 1126081 309763 215030 at GRSFI Standard 11.17 1118736 96731 38340 at HIP1R Standard 11.26 1124613 296720 212599 at AUTS2 Standard 11.43 ;1125456 300592 213906 at MYBL1 Standard 11.60 1097177 9691 224761 at GNA13 Standard 12.11 1120400 152207 203787 at SSBP2 Standard 12.12 1139266 76640 218723 s at RGC32 Standard 12.22 1100770 65578 228976 at Standard 12.73 1131246 153752 201853 s at CDC25B Standard 13.48 1096503 21379 223522 at C9orf45 Standard 14.50 1124920 6150 213039 at ARHGEF1 Standard 15.03 1128360 445043 218988 at SLC35E3 Standard 15.24 1099444 434489 227407 at FLJ90013 Standard 21.03 1134582 78202 208794 s at SMARCA4 Standard Mean ABC -4179.76 Cut 1 0.20 Mean BL -1894.68 Cut 2 0.80 Covarlance ABC 53707.58 Covariance BL 194887.5 Table 45: ABC vs. GCB Signature Scale UNIQID Unigene ID Build 167 Probe set Gene symbol http://www.ncbi.nlm.nih. gov/UniGene Standard -15.31 1122645 158341 207641 at TNFRSF13B Standard -14.56 1120651 80205 204269 at PIM2 Standard -14.18 1120808 127686 204562 at IRF4 Standard -13.84 1114824 193370 222762 x at LIMD1 Standard -13.44 1136687 59943 212345 s at CREB3L2 Standard -13.12 1139805 414362 220230 s at CYB5R2 Standard -12.23 1104552 193857 233483 at LOC96597 147 Standard -12.19 1097236 235860 224837 at FOXP1 Standard -12.06 1121629 41691 205965at BATF Standard -11.93 1128195 115325 218699 at RAB7LI Standard -11.72 1111503 502910 241383 at KBRAS2 Standard -11.66 1134991 444105 209474 s at ENTPD1 Standard -11.27 1098678 386140 226530 at BMF Standard -10.9 1131074 76894 201572 _xat DCTD Standard -10.82 1135165 170359 209827 s at IL16 Standard -10.7 1132396 118722 203988 s at FUT8 Standard -10.54 1131541 310230 202369 s at TRAM2 Standard -10.47 1105759 171262 235056 at ETV6 Standard -10.38 1121564 437783 205865 at ARID3A Standard -10.16 1130472 192374 200599 s at TRA1 Standard -10.04 1132058 161999 203313 s at TGIF Standard -10.03 1105684 195155 234973 at SLC38A5 Standard -9.95 1097735 26765 225436 at LOC58489-. Standard -9.94 1115071 390476 223218 s at MAIL Standard -9.85 1101149 517226 229437 at -BC Standard -9.83 1119884 418004 202716 at PTPN1 Standard -9.71 1134095 89555 208018 s at HCK Standard -9.68 1135550 221811 210550 s at RASGRF1 Standard -9.61 1098927 356216 226811 at FLJ20202 Standard -9.6 1120389 75367 203761 at SLA Standard -9.58 1133910 167746 207655 s at BLNK Standard 9.56 1118736 96731 38340 at HIP1R Standard 9.58 1128860 323634 219753 at STAG3 Standard 9.68 1134582 78202 208794 s at SMARCA4 Standard 9.7 1121853 98243 206310 at SPINK2 Standard 10.14 1119258 88556 201209 at HDACI Standard 10.19 1132122 307734 203434 s at MME Standard 10.23 1120400 152207 203787 at SSBP2 Standard 10.48 1529344 317970 Lymph Dx_ 065 at SERPINAI I Standard 10.64 1124613 296720 212599 at AUTS2 Standard 10.72 1132159 147868 203521 s at ZNF318 Standard 10.98 1097901 266175 225626_at PAG Standard 11.1 1128287 300063 218862 at ASB13 Standard 12.26 1099686 117721 227684 at . Standard 12.45 1112674 310320 242794 at MAML3 Standard 13.15 1120370 78877 203723 at ITPKB Standard 14.23 1125456 300592 213906 at MYBLI Lymph Node 6.8 1097202 386779 224796 at EDEFI Lymph Node 6.85 1131755 241257_ 202729_s at LTBPI Lymph Node 7.27 1136273 13775 211597 s at HOP Lymph Node 7.35 1119424 75485 201599 at OAT Lymph Node 7.86 1095985 83883 222450 at TMEPA Lymph Node 8.02 1124875 18166 212975 at KIAA0870 Lymph Node 8.32 1124655 79299 212658 at LHFPL2 Lymph Node 8.62 -1115034 387222 223158 s at NEK6 Proliferation -9.11 1120583 153768 204133 at RNU3IP2 Proliferation -7.87 1135492 408615 210448 s at P2RX5 Proliferation -7.68 1127756 313544 217850 at NS Proliferation -7.57 1097195 149931 224785_at MGC29814 148 Proliferation -7.31 1127813 14317 217962_at NOLA3 Proliferation -7.24 1138944 84753 218051 s at FLJ12442 Proliferation -6.99 1139226 266514 218633 x at FLJ11342 Proliferation -6.7 1137486 441069 214442 s at MIZI Proliferation -6.51 1133786 153591 207396 s at ALG3 Proliferation -6.45 1131150 75514 201695 s at NP Proliferation -6.45 1119076 268849 200681 at GLOI Proliferation -6.38 1115679 8345 224523 s at MGC4308 Proliferation -6.34 1110223 212709 239973 at Proliferation -6.3 1529338 284275 LymphDx_058_s_a PAK2 Proliferation -6.24 1135164 458360 209825 s at UMPK Proliferation -6.24 1128738 335550 219581 at MGC2776 Proliferation -6.01 1099088 14355 226996 at Proliferation -5.98 '1123192 315177 209100 at IFRD2 Proliferation -5.83 1116073 146161 227103 s at MGC2408 Proliferation 5.79 1097388 278839 225024 at C20orf77 Proliferation 6.13 1124563 249441 212533 at WEEI Standard Lymph Node Proliferation Mean ABC -2226.57 476.67 -1096.34 Cut 1 0.50 Mean GCB -1352.02 547.18 -1005.72 Cut 2 0.74 Covarlance ABC 33472.10 3418.91 4347.99 3418.91 1296.05 846.32 4347.99 846.32 1609.13 Covariance GCB 53751.59 466.34 - 751.08 466.34 777.74 249.29 751.08 249.29 1708.67 1 Table 46: ABC vs. PMBL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -14.61 1097236 235860 224837 at FOXPI Standard -14.47 1104552 193857 233483 at LOC96597 Standard -13.62 1122645 158341 207641 at TNFRSF13B Standard -12.05 1135102 349845 209685 s at PRKCBI Standard -11.65 1096499 293867 223514 at CARDI I Standard -11.26 1124770 153261 212827 at IGHM Standard -11.25 1125010 43728 213170 at GPX7 Standard -11.13 1109545 63187 239231 at Standard -10.99 1109220 445977 238880 at GTF3A Standard -10.87 1131074 76894 201572 x at DCTD Standard -10.68 1134517 75807 208690 s at PDLIMI Standard -10.63 1098604 32793 226444_at SLC39AI0 Standard -10.56 1131219 109150 201810 s at SH3BPS Standard -10.52 1120651 80205 204269 at PIM2 Standard -10.39 1133910 167746 207655 s at BLNK Standard -10.32 1099396 435949 227346 at ZNFNIAI Standard -10.25 1529297 132335 Lymph_Dx 01 5 at 149 Standard -10.17 1107575 424589 237033 at MGC52498 Standard -10.11 1117211 356509 233955 x at HSPC195 Standard 10.06 1129517 -33 220712 at Standard 10.29 1139950 437385 220731 s at FLJ10420 Standard 10.35 1097553 197071 225214 at PSMB7 Standard 10.41 1119516 6061 201834 at PRKAB1 Standard 10.47 1122772 66742 207900 at CCL17 Standard 10.55 1132762 80395 204777 s at MAL Standard 10.77 1099265 375762 227193 at Standard 10.81 1095996 288801 222482 at SSBP3 Standard 11.14 1100770 65578 228976 at Standard 11.19 1133801 181097 207426 s at TNFSF4 Standard 11.61 1099154 97927 227066 at MOBKL2C Standard 11.63 1120370 78877 203723 at ITPKB Standard 11.8 1112674 310320 242794 at MAML3 Standard 12.57 1105178 283961 234284 at GNG8 Standard 12.63 1124613 296720 212599 at AUTS2 Standard 13.28 1106415 169071 235774 at Standard 13.3 1121762 32970 206181 at SLAMFI Standard 13.6 1121853 .98243 206310 at SPINK2 Lymph Node 10.91 1105838 129837 235142 at ZBTB8 Lymph Node 10.99 1136273 13775 211597 s at HOP Lymph Node 11.02 1099418 172792 227370 at KIAA1946 Lymph Node 11.46 1124875 18166 212975 at KIAA0870 Lymph Node 11.99 1120299 79334 203574 at NFIL3 Lymph Node 12.49 1135871 104717 211031 s at CYLN2 Lymph Node 13.33 1121767 458324 206187 at PTGIR Proliferation -13.17 1138944 84753 218051 s at FLJ12442 Proliferation -11.61 1116122 42768 227408 s at DKFZp76100113 Proliferation -11.16 1110223 212709 239973 at Proliferation -9.93 1120717 444159 204394 at SLC43AI Proliferation -9.54 1110099 116665 239835 at TA-KRP Proliferation -9.49 1130942 445977 201338 x at GTF3A Proliferation -9.28 1123192 315177 209100 at IFRD2 Proliferation -9.14 1135492 408615 210448 s at P2RX5 Proliferation -9.03 1120011 3068 202983 at SMARCA3 Proliferation -9.01 1096738 87968 223903 at TLR9 Proliferation -8.91 1108961 292088 238593 at FLJ22531' Standard Lymph Node Proliferation Mean ABC -849.47 531.79 -1027.48 Cut 1I 0.20 Mean PMBL 27.99 750.84 -872.43 Cut 2 0.8O Covariance ABC 14028.46 3705.84 3118.60 3705.84 2326.91 1083.37 3118.60 1083.37 1589.42 Covariance PMBL 19425.29 5109.98 2199.28 5109.98 2084.28 620.86 1- 2199.28 620.86 1028.44 Table 47: BL vs. GCB Signature Scale UNIQI gen D Build 7 Probe set Gene Symbo 150 nih.gov/UniGene ' Standard -12.78 1131246 153752 201853 s at CDC25B Standard -11.35 1099444 434489 227407 at FLJ90013 Standard -10.4 1116432 409362 229356 x at KIAA1259 Standard -10.3 1134582 78202 208794 s at SMARCA4 Standard -10.01 1133998 76884 207826 s at ID3 Standard -9.3 1126081 309763 215030 at GRSF1 Standard -9.19 1096503 21379 223522 at C9orf45 Standard -8.95 1529340 -99 LymphDx 06 Tf3-8 12-8 1 at Standard -8.88 1138128 390428 216199 s at MAP3K4 Standard -8.8 1099152 351247 227064 at MGC15396 Standard -8.69 1133757 6113 207320._x_at STAU Standard -8.54 1116593 422889 230329 s at NUDT6 Standard -8.4 1130926 508741 201310 s at C5orf13 Standard -8.39 1135685 371282 210776 x at TCF3 Standard -8.39 1140520 11747 221741 s at C20orf2l Standard -8.34 1119802 7370 202522 at PITPNB Standard -8.31 1096149 410205 222824 at NUDT5 Standard -8.23 1124786 . 22370 212847 at NEXN Standard -8.07 1098012 355669 225756 at CSNK1 E Standard -7.89 1116317 526415 228661 s at Standard -7.86 1109195 416155 238853 at Standard -7.71 1134880 168799 209265 s at METTL3 Standard -7.66 1529298 136707 Lymph Dx 01 6 at Standard -7.55 1128660 413071 219471 at C13orf18 Standard -7.55 1138973 11270 218097 s at ClOorf66 Standard -7.46 1127294 421986 217028 at CXCR4 Standard 7.47 1134270 352119 208284 x at GGT1 Standard 7.48 1120743 79197 204440 at CD83 Standard 7.5 1098179 163725 225956 at LOC153222 Standard 7.55 1121400 223474 205599 at TRAFI Standard 7.59 1114967 7905 223028 s at SNX9 Standard 7.6 1122087 72927 206693 at IL7 Standard 7.64 1101905 170843 230345 at Standard 7.77 1120700 410745 204362 at SCAP2 Standard 7.8 1120572 84 204116 at IL2RG Standard 7.84 1098271 300670 226056 at CDGAP Standard 7.9 1115073 131315 223220 s at BAL Standard 7.9 1133210 434374 205842 s at JAK2 Standard 8 1129269 62919 220358 at SNFT Standard 8.01 1131940 1103 203085 s at TGFBI Standard 8.07 1098506 193400 226333 at IL6R Standard 8.13 1120601 441129 204166 at KIAA0963 Standard 8.21 1102540 434881 231093 at FCRH3 Standard 8.24 1121695 511759 206082 at HCP5 Standard 8.33 1136877 409934 212998 x at HLA-DQB1 Standard 8.37 1100138 278391 228234 at TIRP Standard 8.46 1126293 504816 215346 at TNFRSF5 Standard 8.46 1127805 380627 217947 at CKLFSF6 Standard 8.59 1136573 914 211991 s at HLA-DPAI 15 1 Standard 8.62 1119111 35052 200804 at TEGT Standard 8.7 1136329 132739 211675 s at HIC Standard 8.74 1123690 111805 210176 at TLRI Standard 8.81 1138677 390440 217436 x at Standard 8.89 1113993 131811 244286 at Standard 8.89 1132651 439767 204529 s at TOX Standard 8.91 1119566 433506 201954_at ARPC1B Standard 9.01 1128626 501452 219424 at EB13 Standard 9.17 1101272 179089 229584 at DKFZp434H2111 Standard 9.33 1136777 387679 212671 s at HLA-DQA1 Standard 9.33 1109756 530304 239453 at Standard 9.4 1136216 512152 211528 x at HLA-G Standard 9.4 1124381 440808 212288 at FNBPI Standard 9.46 1099680 210387 227677 at JAK3 Standard 9.49 1109913 355724 239629 at CFLAR Standard 9.55 1132636 306278 204490 s at CD44 Standard 9.59 1119243 440165 201171_at ATP6VOE Standard 9.72 1101149 517226 229437 at BIC Standard 9.8 1130674 381008 200905 x at HLA-E Standard 10.34 1119939 170087 202820 at AHR Standard 10.44 1132883 432453 205027_s at MAP3K8 Standard 10.74 1121452 227817 205681 at BCL2A1 Standard 10.84 1137360 429658 214196 s at CLN2 Standard 12.08 1132520 283063 204249 s at LMO2 Standard 12.33 1131497 114931 202295 s at CTSH Standard 13.58 1123163 421342 208991 at STAT3 Lymph Node -9.1 1138136 433574 216215_sat RBM9 Lymph Node 8.78 1130121 411958 221978 at HLA-F Lymph Node 9.22 1139830 221851 220330 s at SAMSN1 Lymph Node 9.23 1131705 386467 202638 s at ICAMI Lymph Node 9.62 1130168 75626 222061 at CD58 Lymph Node 9.66 1121844 83077 206295 at IL18 Lymph Node 9.68 1121000 519033 204924 at TLR2 Lymph Node 9.83 1102437 437023 230966 at IL411 Lymph Node 10.71 1119475 296323 201739 at SGK Lymph Node 11.09 1131786 375957 202803 s at ITGB2 Proliferation -11.07 1133141 344524 205677 s at DLEUI Proliferation -10.04 1138259 89525 216484 x at HDGF Proliferation -9.74 1131578 202453 202431 s at MYC Proliferation -9.45 1137449 223745 214363 s at MATR3 Proliferation -9.43 1130468 166463 200594 x at HNRPU Proliferation -9.21 1138157 82563 216251 s at KIAA0153 Proliferation -9.15 1127756 313544 217850 at NS Proliferation -9 1130433 246112 200058 s at U5-200KD Proliferation -8.76 1123108 108112 208828 at POLE3 Proliferation -8.75 1128738 335550 219581 at MGC2776 Proliferation -8.74 1122400 439911 207199 at TERT Proliferation -8.66 1097948 69476 225684 at LOC348235 Proliferation -8.6 1119460 76122 201696 at SFRS4 Proliferation -8.6 1136401 27258 211761 s at SIP Proliferation -8.58 1099088 14355 226996 at Proliferation -8.51 1134653 253536 208901 s at TOP1 152 Proliferation -8.49 .1140584 294083 221932_s at C14orf87 Proliferation -8.43 1121309 23642 205449 at HSU79266 Proliferation -8.43 1120385 36708 203755 at BUB1B Proliferation -8.38 1136710 75782 212429_s at GTF3C2 Proliferation -8.36 1136605 448398 212064 x at MAZ Proliferation -8.24 1120697 323462 204355 at DHX30 Proliferation -8.19 1127833 382044 218001 at MRPS2 Proliferation -8.11 1096903 437460 224185 at FLJ10385 Proliferation -8.1 1120596 4854 204159 at CDKN2C Proliferation -8.1 1120779 28853 204510_at CDC7 Standard Lymph Node Proliferation Mean BL 1098.69 576.05 -2392.12 Cut 1 0.09 Mean GCB 2187.37 768.53 -2129.35 Cut2 0.53 Covariance BL 75263.67 12684.43 15734.77 12684.43 2650.81 2358.05 15734.77 2358.05 4653.00 Covariance GCB 50548.22 9301.12 14182.83 9301.12 2602.51 3028.21 14182.83 3028.21 5983.04 Table 48: BL vs. PMBL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -13.54 1099444 434489 227407 at FLJ90013 Standard -13.42 1096503 21379 223522 at C9orf45 Standard -13.36 1130114 445084 221965 at MPHOSPH9 Standard . -13.27 1124786 22370 212847_at NEXN Standard -13.27 1134582 78202 208794 s at SMARCA4 Standard -12.37 1096149 410205 222824 at NUDT5 Standard -11.95 1130855 77515 201189 s at ITPR3 Standard -11.66 1529298 136707 Lymph_Dx_016 at Standard -11.35 1131246 153752 201853 s at CDC25B Standard -11.17 1136925 436939 213154 s at BICD2 Standard -11.08 1124188 282346 211987 at TOP2B Standard -11.06 1133998 76884 207826 s at ID3 Standard -10.76 1139266 76640 218723 s at RGC32 Standard -10.74 1134880 168799 209265 s at METTL3 Standard -10.69 1140520 11747 221741 s at C20orf21 Standard -10.6 1109545 63187 239231 at Standard -10.55 1106043 266331 235372 at FREB Standard -10.52 1110214 144519 239964 at TCL6 Standard -10.49 1098592 283707 226431 at ALS2CR13 Standard -10.45 1109220 445977 2 38880 at GTF3A Standard -10.41 1131263 249955 201877 s at PPP2R5C Standard 10.54 1122772 66742 207900 at CCL17 Standard 10.59 1109913 355724 239629 at CFLAR Standard 10.82 1119884 418004 202716 at PTPNI 153 Standard 10.83 1135189 137569 209863 s at TP73L Standard 10.89 1123437 73090 209636 at NFKB2 Standard 11.15 1124381 440808 212288_at FNBPI Standard 11.26 1108237 126232 237753_at Standard 11.34 1101149 517226 229437 at BIC Standard 11.77 1139774 15827 220140 s at SNX11 Standard 11.87 1123163 421342 208991 at STAT3 Standard 11.93 1129269 62919 220358_at SNFT Standard 12.03 1132636 306278 204490 s at CD44 Standard 12.1 1138677 390440 217436 x at Standard 12.2 1139950 437385 220731 s at FLJ10420 Standard 12.25 1134270 352119 208284 x at GGT1 Standard 12.27 1136216 512152 211528 x at HLA-G Standard 12.79 1121400 223474 205599 at TRAF1 Standard 12.82 1119939 170087 202820 at AHR Standard 13.12 1126293 504816 215346 at TNFRSF5 Standard 13:44 1100138 278391 228234 at TIRP Standard 13.74 1132883 432453 205027 s at MAP3K8 Standard 13.94 1131497 114931 202295 s at CTSH Standard 14.15 1121762 32970 206181 at SLAMF1 Standard 14.51 1132520 283063 204249 s at LMO2 Standard 14.68 1121452 227817 205681 at BCL2A1 Standard 15.24 1105178 283961 234284 at GNG8 Lymph Node 10.95 1121205 2488 205269 at LCP2 Lymph Node 11.22 1140845 21486 AFFX-- STAT1 HUMISGF3A/M 97935 3 at Lymph Node 11.45 1131068 118400 201564 s at FSCN1 Lymph Node 11.92 1131705 386467 202638 s at ICAM1 Lymph Node 12.06 1131038 81328 201502 s at NFKBIA Lymph Node 12.49 1121444 153563 205668 at LY75 Lymph Node 13.01 1123457 446304 209684 at RIN2 Lymph Node 13.19 1140404 354740 221584 s at KCNMA1 Lymph Node 13.26 1124875 18166 212975 at KIAA0870 Lymph Node 14.06 1102437 437023 230966 at IL411 Lymph Node 14.11 1132766 82359 204781 s at TNFRSF6 Lymph Node 15.31 1121767 458324 206187 at PTGIR Lymph Node 15.32 1135871 104717 211031 s at CYLN2 Lymph Node 15.34 1138652 444471 217388 s at KYNU Lymph Node 16.01 1139830 221851 220330 _s at SAMSN1 Standard Lymph Node Mean BL -66.97 1445.63 Cut 1 0.20 Mean PMBL 1205.38 2041.25 Cut 2 0.80 Covariance BL 35263.67 13424.88 13424.88 7458.56 Covariance PMBL 12064.38 5113.74 5113.74 3216.53 Table 49: FH vs. DLBCL-BL Signature | Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol 154 http://www.ncbi.nim. nih.gov/UniGene Standard -12.81 1104910 458262 233969 at IGL@ Standard -11.54 1102898 145519 231496_at FKSG87 Standard -11.46 1117298 449586 234366_x_at Standard -11.46 1132973 169294 205255 x at TCF7 Standard -11.22 1133099 88646 205554 s at DNASE1L3 Standard -10.76 1131531 153647 202350 s at MATN2 Standard -10.59 1124283 406612 212144 at UNC84B Standard -10.35 1099847 36723 227867_at LOC129293 Standard -10.22 1136430 102950 211798 x at IGLJ3 Standard -10.05 1117394 -13 234792 x at Standard -9.95 1133047 528338 205434 s at AAKI Standard -9.95 1098865 250905 226741 at LOC51234 Standard -9.82 1108515 98132 238071 at LCN6 Standard -9.8 1131407 154248 202125 s at ALS2CR3 Standard -9.77 1128469 390817 219173 at FLJ22686 Standard -9.7 1123875 428 210607 at FLT3LG Standard -9.69 1131875 169172 202965 s at CAPN6 Standard -9.69 1135173 3781 209841 s at LRRN3 Standard -9.48 1099798 411081 227811 at FGD3 Standard -9.41 1119046 349499 200606 at DSP Standard -9.36 1122449 278694 207277 at CD209 Standard -9.34 1114017 133255 244313 at Standard -9.34 1122767 652 207892_at TNFSF5 Standard -9.24 1123369 79025 209481 at SNRK Standard -9.16 1098954 128905 226844 at MOBKL2B Standard -9.14 1135513 421437 210481 _s at CD209L Standard -9.08 1100904 426296 229145 at LOC119504 Standard -8.99 1122738 81743 207840 at CD160 Standard -8.94 1120925 204891 204773 at ILIIRA Standard 9.09 1123055 185726 208691 at TFRC Standard 9.62 1134858 405954 209226 s at TNPO1 Standard 10.19 1123052 180909 208680 at PRDX1 Standard 10.81 1124178 446579 211969 at HSPCA Lymph Node -10.59 1137597 3903 214721 x at CDC42EP4 Lymph Node -9.69 1119684 439586 202242 at TM4SF2 Lymph Node -9.25 1125593 8910 214180 at MANICI Lymph Node -8.44 1124318 21858 212190 at SERPINE2 Lymph Node -8.09 1119448 212296 201656_at ITGA6 Lymph Node -8.07 1125546 125036 214081 at PLXDC1 Lymph Node -7.7 1097683 132569 225373_at PP2135 Lymph Node -7.56 1101305 112742 229623_at Lymph Node 7.45 1135240 436852 209955_s_at FAP Proliferation 6.97 1135101 20830 209680 s at KIFC1 Proliferation 7.03 1130426 432607 200039 s at PSMB2 Proliferation 7.04 1130501 2795 200650 s at LDHA Proliferation 7.08 1130744 158688 201027 s at EIF5B Proliferation 7.23 1137506 75258 214501 s at H2AFY Proliferation 7.32 1131474 95577 202246 s at CDK4 Proliferation 7.39 1130871 159087 201222 s at RAD23B Proliferation 7.42 1119375 381072 201489 at PPIF 155 Proliferation 7.47 1136595 404814 212038 s _at VDAC1 Proliferation 7.7 1135858 90093 211015 s at HSPA4 Proliferation 7.78 1130527 184233 200692 s at HSPA9B Proliferation 7.78 1130820 151777 201144 s at EIF2S1 Proliferation 7.83 1115829 433213 225253 s at METTL2 Proliferation 7.84 1134699 439683 208974 x at KPNBI Proliferation 7.87 1120274 31584 203517 at MTX2 Proliferation 7.92 1136786 63788 212694 s at PCCB Proliferation 7.95 1097172 434886 224753 at CDCA5 Proliferation 8.4 1138537 -12 217140 s at Proliferation 8.53 1119488 154672 201761 at MTHFD2 Proliferation 8.58 1130799 233952 201114 x at PSMA7 Proliferation 8.72 1135673 82159 210759 s at PSMA1 Proliferation 9.4 1114679 16470 222503 s_ at FLJ10904 Standard Lymph Node Proliferation Mean FH -2193.59 -588.21 1571.78 Cut 1 0.50 Mean DLBCL-BL -1448.27 -441.91 1735.00 Cut2 0.92 Covariance FH 6729.73 1223.99 2541.22 1223.99 405.22 293.72 2541.22 293.72 1797.58 Covariance DLBCL-BL 17675.23 3642.41 4158.43 3642.41 1379.81 1066.48 4158.43 1066.48 2858.21 Table 50: FH vs. FL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -11.23 1117298 449586 234366 x at Standard -10.62 1121953 38365 206478 at KIAA0125 Standard -10.6 1104910 458262 233969 at IGL@ Standard -10.39 1136430 102950 211798 x at IGLJ3 Standard -9.96 1129281 395486 220377 at C14orfl 10 Standard -9.73 1118835 102336 47069 at ARHGAP8 Standard -9.21 1127807 7236 217950_at NOSIP Standard -9.05 1128377 371003 219014 at- PLAC8 Standard -8.85 1101004 2969 229265 at SKI Standard 9.06 1139411 368238 219073 s at OSBPL10 Standard 9.07 1120789 154729 204524 at PDPK1 Standard 9.21 1136464 159428 211833 s at BAX Standard 9.29 1125279 445652 213575_at TRA2A Standard 9.45 1529390 79241 LymphDx 12 BCL2 0 at Standard 9.52 1132022 173911 203247 s at ZNF24 Standard 9.57 1139645 134051 219757 s at C14orf1O Standard 9.64 1137561 67397 214639 s at HOXA1 Standard 9.66 1114893 314623 222891 s at BCL11A Standard 10.38 1098095 131059 225852 at ANKRDI7 Standard 10.4 1134858 405954 209226 s at TNPO1 Standard 12.65 1101054 173328 229322 at PPP2R5E Standard 12.79 1124178 446579 211969 at HSPCA 156 Standard 13.34 | 1135489 288178 210438_x at SSA2 Standard Mean FH 136.43 Cut 1 0.50 Mean FL 640.38 Cut 2 0.99 Covariance FH 10719.40 Covariance FL 9373.11 Table 51: FH vs. MCL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http:llwww.ncbi.nlm. nih.gov/UniGene Standard 13.05 1100258 88442 228377 at KIAA1 384 Standard 13.43 1529382 371468 Lymph Dx 11 CCND1 1 at Standard 13.54 1106855 455101 236255 at KIAA1909 Standard 13.73 1529308 193014 LymphDx 02 7 x at Standard 14.56 1100873 445884 229103 at Standard 21.12 1132834 432638 204914 s at SOXi1 Lymph Node -8.44 1130378 234434 44783_s at HEY1 Lymph Node -7.92 1123552 423077 209879 at SELPLG Lymph Node -7.7 1131218 76753 201809 s at ENG Lymph Node -7.4 1097683 132569 225373 at PP2135 Lymph Node -7.15 1136273 13775 211597 s at HOP Lymph Node 14.16 1134532 371468 208711 s at CCND1 Standard Lymph Node Mean FH 451.68 -282.65 Cut 1 0.20 Mean MCFL 863.16 -156.82 Cut 2 0.80 Covariance FH 1617.92 222.89 222.89 271.65 Covariance MCL 3154.38 917.30 .917.30 659.94 Table 52: FH vs. SLL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -13.14 1120765 343329 204484 at PIK3C2B Standard -12.9 1097897 266175 225622 at PAG Standard 12.72 1133195 274243 205805 s at ROR1 Standard 12.74 1140416 58831 221601_s at TOSO Standard 13.53 1131687 369280 202606 s at TLK1 Standard 13.57 1107044 163426 236458 at Standard 14.43 1529389 79241 LymphDx 119 BCL2 Sat Standard 14.51 1129026 135146 220007_at FLJ13984 Standard 14.77 1136987 21695 213370 s at SFMBT1 Standard 14.79 1137109 469653 213689 x at RPL5 Standard 15.37 1529308 193014 Lymph Dx_027 ____ ___ ___ __ ___ __ ___ ___ ___ ___ ___ ___ ___x._at 157 Standard 15.82 1120832 57856 204604 at PFTK1 Standard 17.37 1135550 221811 210550 s at RASGRF1 Standard 18.98 1122864 434384 208195_at TTN Lymph Node -12.89 1123038 119000 208636 at ACTN1 Lymph Node -12.8 1130378 234434 44783 s at HEYI Lymph Node -11.59 1124875 18166 212975 at KIAA0870 Lymph Node -11.47 1103497 50115 232231 at Lymph Node -10.31 1099358 93135 227300 at Lymph Node -10.27 1121129 285401 205159_at CSF2RB Lymph Node -10.23 1100249 388674 228367_at HAK Lymph Node -10.05 1132345 109225 203868_ s at VCAM1 Lymph Node -9.93 1123401 50130 209550_at NDN Lymph Node -9.75 1120500 82568 203979 at CYP27A1 Lymph Node -9.57 1124318 21858 212190 at SERPINE2 Lymph Node -9.48 1120288 17483 203547 at CD4 Lymph Node -9.45 1123372 195825 209487 at RBPMS Lymph Node -9.39 1123376 37682 209496 at RARRES2 Lymph Node -9.29 1123213 12956 209154 at TIP-I Lymph Node -9.23 1098412 409515 226225 at MCC Lymph Node -9.23 1125593 8910 214180 at MAN1C1 Lymph Node -9.17 113178Q 375957 202803_s at ITGB2 Lymph Node -9.04 1097683 132569 225373 at PP2135 Lymph Node -8.91 1097255 380144 224861 at Lymph Node -8.76 1131068 118400 201564 s at FSCN1 Lymph Node -8.7 1119074 54457 200675 at CD81 Lymph Node -8.68 1125130 35861 213338 at RISI Lymph Node -8.59 1139661 416456 219806 s at FN5 Standard Lymph Node Mean FH 1144.02 -2223.71 Cut 1 0.20 Mean SLL 1592.27 -1798.11 Cut 2 0.80 Covariance FH 902.56 442.69 442.69 809.90 Covariance SLL 2426.26 2938.58 2938.58 9435.72 5 Table 53: FL vs. DLBCL-BL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -23.03 1124833 356416 212914 at CBX7 Standard -22.25 1099204 193784 227121 at Standard -22.2 1119766 93231 202423 at MYST3 Standard -22.04 1099798 411081 227811 at FGD3 Standard -22.01 1102898 145519 231496 at FKSG87 Standard -21.79 1131197 269902 201778 s at KIAA0494 Standard -21.69 1098415 130900 226230 at KIAA1 387 Standard -21.57 1120834 57907 204606 at CCL21 Standard -21.39 -1130155 436657 222043_ at CLU 158 Standard -20.98 1100904 426296 229145 at LOC119504 Standard -20.8 1131531 153647 202350 s at MATN2 Standard -20.72 1137582 433732 214683 s at CLK1 Standard -20.66 1119782 155418 202478 at TRB2 Standard -20.59 1122767 652 207892_at TNFSF5 Standard -20.58 1125001 16193 213158 at Standard -20.56 1134921 413513 209341 s at IKBKB Standard -20.56 1132973 169294 205255_x at TCF7 Standard -20.53 1136984 498154 213364 s at SNXI Standard -20.41 1115888 35096 225629 s at ZBTB4 Standard -20.37 1120160 436976 203288_ at KIAA0355 Standard -20.36 1139054 25726 218263 s at LOC58486 Standard -20.31 1130030 301872 221834 at LONP Standard -20.08 1133024 436987 205383_s at ZNF288 Standard -20.05 1124666 526394 212672 at ATM Standard -19.3 1529397 406557 Lymph _Dx 12 CLK4 7 s at Standard -19.16 1116056 243678 226913 s at SOX8 Standard -19.14 1098433 202577 226250_at Standard -19.1 1123635 408614 210073 at SIAT8A Standard -18.95 1138920 24395 218002 s at CXCL14 Standard -18.84 1133099 88646 205554 s at DNASE1L3 Standard -18.83 1098495 443668 226318 at TBRG1 Standard -18.64 1100879 119983 229111_at MASP2 Standard -18.59 1120695 385685 204352 at TRAF5 Standard -18.55 1119983 409783 202920 at ANK2 Standard -18.5 1101276 1098 229588 at ERdj5 Standard -18.47 1099140 500350 227052 at Standard -18.46 1529331 374126 Lymph Dx 05 i s at Standard -18.45 1131752 170133 202724 s at FOXOIA Standard -18.45 1099265 375762 227193 at Standard -18.32 1098179 163725 225956 at LOC153222 Standard -18.29 1119568 269777 201957 at PPP1R12B Standard -18.19 1099900 444508 227934 at Standard -18.17 1119361 391858 201448 at TIA1 Standard -18.02 1121650 421137 206002 at GPR64 Standard -17.91 1100911 320147 229152 at C4orf7 Standard -17.86 1529285 348929 LymphDx_00 KiAA1219 2 at Standard -17.47 1529357 444651 LymphDx 08 1 at Standard -17.42 1131863 2316 202936 s at SOX9 Standard -17.16 1129943 512828 221626 at ZNF506 Standard -17.12 1121301 449971 205437 at ZNF134 Standard -17.11 1131340 437457 202018 s at LTF Standard -17.1 1124606 444324 212588_at PTPRC Standard -17..08 1131407 154248 202125 s at ALS2CR3 Standard -16.97 1118939 198161 60528 at PLA2G4B Standard -16.91 1134738 75842 209033 s at DYRKIA Standard -16.9 1134083 285091 207996 s at C18orfl Standard -16.89 1120925 204891 204773 at IL1IRA Standard -16.86 1110070 -101 239803 at 159 Standard -16.83 1100042 351413 228113_at RAB37 Standard -16.82 1120134 75545 203233 at IL4R Standard -16.75 1124283 406612 212144 at UNC84B Standard -16.72 1109603 -100 239292 at Standard -16.71 1120509 155090 204000_at GNB5 Standard -16.65 1133538 1416 206760 s at FCER2 Standard -16.64 1130735 179526 201009 s at TXNIP Standard -16.59 1100150 9343 228248_at MGC39830 Standard -16.54 1124237 258855 212080 at MLL Standard -16.51 1124416 283604 212331 at RBL2 Standard -16.48 1133091 73792 205544 s at CR2 Standard -16.46 1131263 249955 201877 s at PPP2R5C Standard -16.44 1118347 528404 243366 s at ITGA4 Standard -16.43 1529343 521948 LymphDx_06 4 at Standard -16.43 1099549 446665 227533 at Standard 17.05 1529453 372679 LymphDx 08 FCGR3A 5 at Standard 17.41 1097540 388087 225195 at Standard 18.47 1140473 17377 221676 s at COROIC Standard 18.55 1121100 301921 205098 at CCRI Standard 20.07 1124254 301743 212110 at SLC39A14 Standard 20.2 1130771 61153 201068 s at PSMC2 Standard 21.46 1137583 273415 214687 x at ALDOA Standard 21.55 1098168 22151 225943 at NLN Standard 24.07 1123055 185726 208691 at TFRC Standard 24.09 1123052 180909 208680 at PRDX1 Lymph Node -20.5 1137597 3903 214721 x at CDC42EP4 Lymph Node -18.52 1124318 21858 212190 at SERPINE2 Lymph Node -18.5 1136762 380138 212624 s at CHN1 Lymph Node -18.07 1101305 112742 229623 at Lymph Node -17.75 1100249 388674 228367 at HAK Lymph Node -16.1 1098412 409515 226225 at MCC Lymph Node -15.51 1140464 111676 221667 s at HSPB8 Lymph Node -15.43 1136832 434959 212842 x at RANBP2L1 Lymph Node -15.37 1119684 439586 202242 at TM4SF2 Lymph Node -15.02 1097448 250607 225093_at UTRN Lymph Node -14.83 1136844 16007 212875 s at C21orf25 Lymph Node -14.73 1135056 169946 209604 s.at GATA3 Lymph Node -14.48 1097202 386779 224796 at DDEFI Lymph Node -14.44 1121278 21355 205399 at DCAMKL1 Lymph Node -14.22 1125009 27621 213169 at Lymph Node -13.97 1100288 26981 228411 at ALS2CRI9 Lymph Node -13.51 1132462 14845 204131_s_at FOXO3A Lymph Node -13.37 1135322 450230 210095 s at IGFBP3 Lymph Node -13.35 1097280 423523 224891_at Lymph Node -12.86 1137097 20107 213656 s at KNS2 Lymph Node -12.85 1098809 359394 226682 at Lymph Node -12.28 1124875 18166 212975_at KIAA0870 Lymph Node -12.18 1132345 109225 203868_s_at VCAM1 Lymph Node -12 1097561 19221 225224_at DKFZP566G142 4 Lymph Node -11.71 1123401 50130 209550_at NDN 160 Lymph Node -11.04 1136996 283749 213397 x at RNASE4 Lymph Node -10.77 1136788 355455 212698 s at 36778 Lymph Node -10.71 1098822 443452 226695 at PRRX1 Lymph Node -10.63 1134200 90786 208161_s at ABCC3 Lymph Node -10.47 1136427 276506 211795 s at FYB Lymph Node -10.46 1121186 100431 205242 at CXCL13 Lymph Node -10.39 1099332 32433 227272 at Lymph Node -10.39 1098978 124863 226869_at Lymph Node -10.22 1103303 49605 232000_at C9orf52 Lymph Node -10.16 1131325 13313 201990_s_at CREBL2 Lymph Node -10.16 10981'74 274401 225949_at LOC340371 Lymph Node -9.93 1124733 66762 212771 at LOC221061 Lymph Node -9.42 1123372 195825 209487_at RBPMS Lymph Node -9.36 1132220 448805 203632 s.at GPRC5B Lymph Node -9.29 1120703 83974 204368 at SLCO2A1 Lymph Node -9.26 1132013 434961 203232 s at SCA1 Lymph Node -9.25 1097307 379754 224929 at LOC340061 Lymph Node -9.18 1119251 433941 201194 at SEPW1 Lymph Node -9.08 1097609 6093 225283 at ARRDC4 Lymph Node -9.07 1136459 252550 211828_s_at KIAA0551 Lymph Node -8.86 1132775 1027 204803 s at RRAD Lymph Node -8.78 1098946 135121 226834 at ASAM Lymph Node -8.68 1140589 433488 221942_s_at GUCYIA3 Lymph Node -8.44 1116966 301124 232744 x at Lymph Node -8.39 1100130 76494 228224_at PRELP Lymph Node -8.36 1110019 -94 239744 at Lymph Node -8.3 1134647 298654 208892 s at DUSP6 Lymph Node -8.28 1125593 8910 214180_at MAN1C1 Lymph Node 7.97 1134370 1422 208438_s at FGR Lymph Node 8.05 1123566 155935 209906 at C3AR1 Lymph Node 8.09 1131119 349656 201647 s at SCARB2 Lymph Node 8.11 1123586 93841 209948_at KCNMB1 Lymph Node 8.13 1128615 104800 219410 at FLJ10134 Lymph Node 8.21 1097297 166254 224917 at VMP1 Lymph Node 8.23 1120299 79334 203574 at NFIL3 Lymph Node 8.37 1128157 23918 218631_at VIP32 Lymph Node 8.4 1130054 82547 221872_at RARRESI Lymph Node 8.41 1098152 377588 225922_at KIAA1 450 Lymph Node 8.53 1101566 98558 229947 at Lymph Node 8.59 1135251 21486 209969 _s _at STATI Lymph Node 8.84 1099167 381105 227080 at MGC45731 Lymph Node 9.01 1132920 753 205119 s at FPR1 Lymph Node 9.26 1097253 77873 224859_at B7H3 Lymph Node 9.29 1120500 82568 203979 at CYP27A1 Lymph Node 9.36 1131507 172928 202311_s_at COL1A1 Lymph Node 9.38 1096456 82407 223454_at CXCL16 Lymph Node 9.49 1136172 38084 211470_s_at SULTIC1 Lymph Node 10:03 1138244 418138 216442 x -at FN1 Lymph Node 10.34 1134424 -17 208540_x_at S100A14 Lymph Node 10.48 1136152 458436 211434 s at CCRL2 Lymph Node 10.51 1118708 7835 37408 at MRC2 Lymph Node 10.6 1136540 179657 211924 s at PLAUR 161 Lymph Node 10.63 1098278 166017 226066 at MITF Lymph Node 10.76 1119477 163867 201743 at CD14 Lymph Node 10.81 1096429 64896 223405_at NPL Lymph Node 11.58 1123672 67846 210152 at LILRB4 Lymph Node 12 1096364 29444 223276 at NID67 Lymph Node 12.16 1119070 445570 200663 at CD63 Lymph Node 12.3 1133065 77274 205479 s at PLAU Lymph Node 12.5 1135240 436852 209955 s at FAP Lymph Node 13.09 1116826 26204 231823 s at KIAA1295 Lymph Node 13.32 1119068 417004 200660 at S10OAll Lymph Node 13.45 1120266 246381 203507 at CD68 Lymph Node 13.63 1133216 502577 205872_x_at PDE4DIP Lymph Node 13.67 1131815 386678 202856 s at SLC16A3 Lymph Node 14.38 1132132 279910 203454 s at ATOX1 Lymph Node 15.25 1134682 411701 208949 s at LGALS3 Lymph Node 15.46 1119237 389964 201141 at GPNMB Lymph Node 15.89 1137698 442669 215001 s at GLUL Lymph Node 17.8 1137782 384944 215223 s at SOD2 Lymph Node 20.11 1130629 135226 200839 s at CTSB Proliferation 21.02 1119375 381072 201489 at PPIF Proliferation 21.24 1119488 154672 201761 at MTHFD2 Proliferation 21.31 1119467 21635 201714 at TUBGI Proliferation 21.68 1130820 151777 201144 s at EIF2S1 Proliferation 21.69 1131474 95577 202246 s at CDK4 Proliferation 22.2 1125249 244723 213523 at CCNE1 Proliferation 22.97 1130501 2795 200650 s at LDHA Proliferation 23.12 1136913 99962 213113_s_at SLC43A3 Proliferation 24.05 1130426 432607 200039 s at PSMB2 Standard Lymph Node Proliferation Mean FL -11121.51 -1603.39 1890.60 Cut 1 0.34 Mean DLBCL-BL -8760.65 -460.71 2101.10 Cut 2 0.94 Covariance FL 246359.77 111505.42 28908.20 111505.42 67036.17 13130.59 28908.20 13130.59 4617.24 Covariance DLBCL-BL 413069.12 178811.32 30151.89 178811.32 106324.53 10877.26 30151.89 10877.26 5180.68 1 Table 54: FL vs. MCL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http:lwww.ncbi.nlm. nih.gov/UniGene Standard -24.56 1123731 17165 210258 at RGS13 Standard -22.56 1133192 24024 205801 s at RASGRP3 Standard -21.12 1114543 156189 244887_at Standard -18.49 1120090 155024 203140_at BCL6 Standard -18.07 1124646 436432 212646_at RAFTLIN Standard -17.24 1132122 307734 203434 s at MME Standard -16.63 1105986 49614 235310 at GCET2 Standard -15.09 1120134 75545 203233 at IL4R 162 Standard -14.05 1132651 439767 204529_s_at TOX Standard 13.8 1098277 6786 226065 at PRICKLE1 Standard 13.85 1109560 207428 239246_at FARP1 Standard 13.86 1103504 142517 232239 at Standard 13.88 1132734 126248 204724 s at COL9A3 Standard 13.91 1115905 301478 225757 s at CLMN Standard 14.89 1098840 55098 226713 at C3orf6 Standard 14.97 1100873 445884 229103 at Standard 14.99 1139393 170129 219032 x at OPN3 Standard 16.13 1124864 411317 212960 at KIAA0882 Standard 16.36 1106855 455101 236255_at KIAA1 909 Standard 16.43 1120858 410683 204647 at HOMER3 Standard 17.38 1130926 508741 201310 s at C5orf13 Standard 18.3 1103711 288718 232478 at Standard 18.62 1109505 8162 239186 at MGC39372 Standard 20.31 1132834 432638 204914_s_at SOXI1 Standard 22.61 1096070 241565 222640_at DNMT3A Standard 28.66 1529382 371468 LymphDx_ 111 CCND1 at Lymph Node -10.77 1097202 386779 224796 at DDEF1 Lymph Node -10.22 1119546 433898 201921_at GNG1O Lymph Node -9.89 1132766 82359 204781 s at TNFRSF6 Lymph Node -9.4 1138867 10706 217892 s at EPLIN Lymph Node 9.65 1125025 301094 213196 at Lymph Node 10.44 1134797 433394 209118 s at TUBA3 Lymph Node 22.6 1529456 371468 LymphDx 113 CCND1 __at Proliferation -7.36 1097948 69476 225684_at LOC348235 Proliferation -7.31 1130747 234489 201030 x at LDHB Proliferation -6.95 1130923 459987 201306 s at ANP32B Proliferation -6.87 1120205 5198 203405 at DSCR2 Proliferation -6.64 1132468 79353 204147 s at TFDP1 Proliferation -6.1 1119916 177584 202780 at OXCT Proliferation -6.08 1119873 446393 202697_at CPSF5 Proliferation -6.08 1119488 154672 201761 at MTHFD2 Proliferation -6.04 1130658 447492 200886_sat PGAM1 Proliferation -5.82 1132825 512813 204900_x_at SAP30 Proliferation -5.53 1115607 435733 224428 s at CDCA7 Proliferation -5.44 1120316 63335 203611_at TERF2 Proliferation -5.34 1114970 279529 223032 x at PX19 Proliferation -5.32 1140843 169476 AFFX- GAPD HUMGAPDH/M 33197 5 at Proliferation -5.28 1131081 180610 201586 s .at SFPQ Proliferation -5.15 1121062 408658 205034 at CCNE2 Proliferation 5.15 1120986 172052 204886_at PLK4 Proliferation 5.16 1097195 149931 224785 at MGC29814 Proliferation 5.2 1120011 3068 202983_at SMARCA3 Proliferation 5.47 _1100183 180582 228286 at FLJ40869 Proliferation 5.67 1121012 96055 204947 at E2FI Proliferation 5.84 1115679 8345 224523_s at MGC4308 Proliferation 5.88 1135285 449501 210024 s at UBE2E3 Proliferation 5.92 1120520 35120 204023 at RFC4 163 Proliferation 6.16 1529361 388681 LymphDx 086 HDAC3 _s at Proliferation 6.45 1096054 21331 222606 at FLJ10036 Proliferation 6.45 1096738 87968 223903 at TLR9 Proliferation 6.51 1136781 120197 212680 x at PPP1R14B Proliferation 6.63 1119466 179718 201710_at MYBL2 Proliferation 6.65 1136285 182490 211615 s at LRPPRC Proliferation 6.67 1136853 66170 212922 s at SMYD2 Proliferation 7.45 1119390 77254 201518 at CBXI Proliferation 8.87 1116122 42768 227408_s_at DKFZp761001 Proliferation 10.12 1119515 3352 201833_at HDAC2 Standard Lymph Node Proliferation Mean FL -18.82 -33.90 23.53 Cut 1 0.14 Mean MCL 1558.10 113.95 165.48 Cut 2 0.58 Covariance FL 21302.14 1098.24 678.04 1098.24 226.29 75.99 678.04 75.99 315.67 Covariance MCL 81008.29 5261.37 9185.20 5261.37 2047.34 875.56 1 9185.20 875.56 1447.43 1 Table 55: FL vs. SLL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -21.04 1123731 17165 210258 at RGS13 Standard. -20.91 1124646 436432 212646 at RAFTLIN Standard -18.82 1099651 120785 227646 at EBF Standard -18.12 1114543 156189 244887 at Standard -17.85 1105986 49614 235310 at GCET2 Standard -16.73 1100911 320147 229152 at C4orf7 Standard -15.77 1132122 307734 203434 s at MME Standard -15.12 1120090 155024 203140 at BCL6 Standard -14.89 1097897 266175 225622 at PAG Standard -14.36 1529343 521948 LymphDx _06 4 at Standard -14.32 1529318 291954 Lymph Dx 03 8 at Standard -14.06 1128694 171466 219517_at ELL3 Standard -13.61 1101586 187884 229971_at GPR114 Standard -13.57 1119752 511745 202391_at BASP1 Standard -13.13 1137561 67397 214639 s at HOXA1 Standard -12.85 1097247 388761 224851 at CDK6 Standard -12.43 1529344 317970 Lymph Dx 06 SERPINA1 1 5at Standard -12.4 1120765 343329 204484 at P(K3C2B Standard -12.33 1130155 436657 222043 at CLU Standard -12.07 1529292 -92 LymphDx 01 0 at Standard -12.01 1119939 170087 202820 at AHR 164 Standard -11.82 1119919 199263 202786 at STK39 Standard -11.77 1099686 117721 227684 at Standard -11.63 1119782 155418 202478 at TRB2 Standard 10.97 1529309 512797 LymphDx 02 HSH2 8 at Standard 10.97 1139393 170129 219032 x at OPN3 Standard 11.04 1131246 153752 201853 s at CDC25B Standard 11.07 1140391 44865 221558 s at LEFI Standard 11.16 1140416 58831 221601 s at TOSO Standard 11.35 1127807 7236 217950 at NOSIP Standard 11.67 1529317 -98 LymphDx 03 7 at Standard 11.81 1117343 306812 234643 x at BUCSI Standard 11.82 1102081 506977 230551 at Standard 11.82 1135042 79015 209582 s at MOX2 Standard 11.96 1132734 126248 204724 s at COL9A3 Standard 12.09 1137109 469653 213689 x at RPL5 Standard 12.14 1099939 488173 227983_at MGC7036 Standard 12.19 1129103 99430 220118 at TZFP Standard 12.47 1135592 758 210621 s at RASAI Standard 12.78 1108970 140489 238604 at Standard 12.92 1097143 74335 224716 at HSPCB Standard 13.18 1136865 412128 212959 s at MGC4170 Standard 13.96 1098220 80720 226002 at GAB1 Standard 14.06 1100847 97411 229070 at C6orflO5 Standard 14.39 1098865 250905 226741 at LOC51234 Standard 15.57 1136687 59943 212345 s at CREB3L2 Standard 15.75 1107044 163426 236458 at Standard 16.52 1123622 8578 210051 at EPAC Standard 17.74 1136987 21695 213370 s at SFMBT1 Standard 19.15 1129026 135146 220007_at FLJ13984 Standard 19.65 1131854 414985 202923_s_at GCLC Lymph Node -14.99 1124875 18166 212975 at KIAA0870 Lymph Node -14.33 1099358 93135 227300 at Lymph Node -13.26 1121129 285401 205159 at CSF2RB Lymph Node -12.61 1119074 54457 200675 at CD81 Lymph Node -12.52 1121029 412999 204971 at CSTA Lymph Node -11.48 1137247 234734 213975 s -at LYZ Lymph Node -10.97 1128781 79741 219648 at FLJ10116 Lymph Node 11.79 1119880 442844 202709 at FMOD Lymph Node 14.4 1134370 1422 208438 s at FGR Standard Lymph Node Mean FL -663.95 -730.08 Cut 1 0.20 Mean SLL 1332.84 -484.93 Cut 2 0.80 Covariance FL 37097.15 1710.73 1710.73 663.78 Covariance SLL 85989.25 17661.52 17661.52 4555.06 165 Table 56: GCB vs. PMBL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -8.39 1096440 231320 223423 at GPR160 Standard -8.13 1096108 292871 222731 at ZDHHC2 Standard -8.12 1125231 446375 213489 at MAPRE2 Standard -8.02 1136759 188882 212605 s at Standard -7.91 1096499 293867 223514 at CARD 11 Standard -7.8 1099388 124024 227336 at DTX1 Standard -7.71 1139623 193736 219667 s at BANKI Standard -7.68 1098592 283707 226431_at ALS2CRI3 Standard -7.67 1107575 424589 237033 at MGC52498 Standard -7.63 1116829 115467 231840 x at LOC90624 Standard -7.42 1130114 445084 221965 at MPHOSPH9 Standard -7.27 1098909 446408 226789 at Standard 7.34 1138759 396404 217707 x at SMARCA2 Standard 7.37 1120355 80420 203687 at CX3CL1 Standard 7.4 1134270 352119 208284 x at GGTI Standard 7.44 1115441 5470 224156 x at ILI7RB Standard 7.78 1103054 341531 231690 at Standard 7.91 1119765 81234 202421 at IGSF3 Standard 7.92 1119438 118110 201641 at BST2 Standard 8.09 1135645 31439 210715 s at SPINT2 Standard 8.15 1106015 96885 235343 at FLJ12505 Standard 8.18 1121400 223474 205599 at TRAF1 Standard 8.38 1139950 437385 220731 s at FLJ10420 Standard 8.73 1122112 1314 206729 at TNFRSF8 Standard 8.77 1122772 66742 207900 at CCL17 Standard 8.84 1132762 80395 204777 s at MAL Standard 9.64 1139774 15827 220140 s at SNXI1 Standard 10.53 1133801 181097 207426 s at TNFSF4 Standard 11.52 1106415 169071 235774 at Standard 12.09 1129269 62919 220358 at SNFT Standard Mean GCB 292.76 Cut 1= 0.16 Mean PMBL 725.28 Cut 2 0.50 Covariance GCB 8538.86 Covariance PMBL 11405.23 5 Table 57: MCL vs. DLBCL-BL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -26.11 1529382 371468 LymphDx 11 CCND1 I at Standard -18.35 1103711 288718 232478 at Standard -17.03 1106855 455101 236255 at KIAA1909 166 Standard -16.49 1098840 55098 226713 at C3orf6 Standard -15.41 1109505 8162 239186 at MGC39372 Standard -15.11 1098954 128905 226844_at MOBKL2B Standard -14.96 1103504 142517 232239 at Standard -14.74 1096070 241565 222640 at DNMT3A Standard -13.81 1137663 247362 214909 s at DDAH2 Standard -13.8 1124864 411317 212960 at KIAA0882 Standard -13.62 1140127 125300 221044 s at TRIM34 Standard -13.62 1119361 391858 201448 at TIAI Standard -13.37 1127849 76691 218032 at SNN Standard 13.72 1133192 24024 205801 s at RASGRP3 Standard 13.85 1137583 273415 214687 x at ALDOA Standard 15.02 1123052 180909 208680 at PRDX1 Standard 16.21 1097611 438993 225285 at BCATI Lymph Node -19.18 1529456 371468 LymphDx 11 CCNDI 3_at Lymph Node -10.71 1098978 124863 226869 at Lymph Node -9.17 1097448 250607 225093_at UTRN Lymph Node 8.84 1135240 436852 209955 s -at FAP Lymph Node 9.11 1119475 296323 201739 at SGK Lymph Node 9.22 1119237 389964 201141 at GPNMB Lymph Node 9.46 1130629 135226 200839 s at CTSB Lymph Node 10.1 1130054 82547 221872 at RARRES1 Standard Lymph Node Mean MCL -1417.55 -25.58 Cut 1 0.50 Mean DLBCL-BL -756.07 202.29 Cut 2 0.88 Covariance MCL 15347.98 3525.48 3525.48 5420.31 Covariance DLBCL-BL 5132.06 1007.64 1007.64 991.38 5 Table 58: MCL vs. SLL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -20.18 1132834 432638 204914 s at SOX1I Standard -15.17 1130926 508741 201310 s at C5orf13 Standard 13.44 1116150 16229 227606 s at AMSH-LP Standard 14.44 1120134 75545 203233 at IL4R Standard 15.18 1529437 445162 LymphDx 17 BTLA _______~~~ ____ __ at Standard 15.19 1529317 -98 LymphDx 03 7 at Standard 16.2 1135042 79015 209582 s at MOX2 Standard Mean MCIL 181.38 Cut 1 0.20 Mean SLL 564.92 Cut 2 0.80 Covariance MCL 1734.42 167 Covariance SLL 910.75 Table 59: SLL vs. DLBCL-BL Signature Scale UNIQID Unigene ID Build 167 Probe set Gene Symbol http://www.ncbi.nlm. nih.gov/UniGene Standard -16.014498 1123622 8578 210051_at EPAC Standard -15.26356533 1102081 506977 230551 at Standard -14.82150028 1107044 163426 236458 at Standard -14.17813266 1098865 250905 226741 at LOC51234 Standard -12.92844719 1110740 416810 240538_at Standard -12.86520757 1129026 135146 220007_at FLJ13984 Standard -12.2702748 1135592 758 210621 s at RASA1 Standard -11.87309449 1117343 306812 234643 x at BUCS1 Standard -11.81789137 1136987 21695 213370 s at SFMBT1 Standard -11.78631706 1124830 9059 212911 at KIAA0962 Standard -11.39454435 1133538 1416 206760 sat FCER2 Standard -11.39050362 1135802 439343 210944_s_at CAPN3' Standard 11.72928644 1120770 300825 204493 at BID Lymph Node -12.21593247 1119880 442844 202709 at FMOD. Lymph Node 9.514704847 1135240 436852 209955 s at FAP Lymph Node 9.739298877 1096429 64896 223405_ at NPL Lymph Node 10.05087645 1119475 296323 201739 at SGK Lymph Node 13.11985922 1119237 389964 201141 at GPNMB Proliferation 10.47525875 1128106 14559 218542 at ClOorf3 Proliferation 10.53295782 1132825 512813 204900 x at SAP30 Proliferation 11.93918891 1130501 2795 200650 s at LDHA Proliferation 11.98738778 1123439 287472 209642 at BUB1 Proliferation 11.99741644 1115607 435733 224428 s at CDCA7 5 Standard Lymph Node Proliferation Mean SLL -1383.640809 177.4452398 467.2463569 Cut 1 0.201266305 Mean DLBCL-BL -926.7275468 329.6795845 582.9070266 Cut 2 0.799816116 Covariance SLL 3591.384775 1789.7516 856.0703202 1789.7516 1421.869535 663.4782048 856.0703202 663.4782048 965.6470151 Covariance DLBCL-BL 2922.643347 473.543487 634.3258773 473.543487 931.9845277 -53.85584619 634.3258773 -53.85584619 767.3545404 As stated above, the foregoing is merely intended to illustrate various embodiments of the present invention. The specific modifications discussed above 10 are not to be construed as limitations on the scope of the invention. It will be apparent to one skilled in the art that various equivalents, changes, and 168 modifications may be made without departing from the scope of the invention, and it is understood that such equivalent embodiments are to be included herein. All references cited herein are incorporated by reference as if fully set forth herein. Abbreviations used herein: ABC, activated B-cell-like diffuse large B cell 5 lymphoma; BL, Burkitt lymphoma; CHOP, cyclophosphamide, doxorubicine, vincristine, and prednisone; Cl, confidence interval; CNS, central nervous system; DLBCL, diffuse large B-cell lymphoma; ECOG, Eastern Cooperative Oncology Group; EST, expressed sequence tag; FACS, fluorescence-activated cell sorting; FH, follicular hyperplasia; FL, follicular lymphoma; GCB, germinal center B-cell-like 0 diffuse large B cell lymphoma; IPI, International Prognostic Index; LPC, lymphoplasmacytic lymphoma; LPS, linear predictor score; MALT, mucosa associated lymphoid tissue lymphomas; MCL, mantle cell lymphoma; MHC, major histocompatibility complex; NA, not available; NK, natural killer; NMZ, nodal marginal zone lymphoma; PCR, polymerase chain reaction; PMBL, primary mediastinal B-cell 5 lymphoma; PTLD, post-transplant lymphoproliferative disorder; REAL, Revised European-American Lymphoma; RPA, RNase protection assay; RR, relative risk of death; RT-PCR, reverse transcriptase polymerase chain reaction; SAGE, serial analysis of gene expression; SLL, small lymphocytic lymphoma; WHO, World Health Organization. !0 Table 2 UNIQlD Probe Set Unigene ID Build Gene 167 Symbol (http://www.ncbi. nlm.nih.gov/UniG ene) 1119003 200004 at 183684 EIF4G2 1119007 200009 at 56845 GD12 1119015 200024 at 378103 RPS5 1130426 200039 s at 432607 PSMB2 1130429 200048 s at 6396 JTB 169 1130430 200052 s at 75117 ILF2 1130433 200058 _s at 246112 U5-200KD 1130446 200076 s at 369785 MGC2749 1130447 200077 s at 446427 OAZ1 1119039 200084 at 447513 SMAP 1130465 200098 s at 7101 ANAPC5 1130468 200594 x at 166463 HNRPU 1130472 200599 s_at 192374 TRAl 1119046 200606 at 349499 DSP 1130482 200616 s at 181418 KIAA0152 ,1130483 200622 x at 334330 CALM3 1119056 200633 at 356190 UBB 1119061 200644 at 75061 MLP 1130501 200650 s at 2795 LDHA 1119068 200660 at 417004 S10OA1l 1119070 200663 at 445570 CD63 1130509 200665 s at 111779 SPARC 1119071 200667 at 411826 UBE2D3 1119072 200670 at 437638 XBP1 1119074 200675 at 54457 C081 1130518 200679 x at 434102 HMGB1 1119076 200681 at 268849 GLO1 1130527 200692 s at 184233 HSPA9B 1130533 200706 s at 76507 LITAF 1119090 200709 at 374638 FKBPIA 1130588 200775 s at 307544 HNRPK 1130603 200797 s at 86386 MCLI 1119111 200804 at 35052 TEGT 1130618 200822 x at 83848 GRCC9 1130622 200829 x at 97128 ZNF207 1130624 200832 s at 119597 SCD 1130629 200839 s at 135226 CTSB 1130631 200842 s at 171292 EPRS 1130645 200860 s at 279949 KIAA1007 1130653 200875 s at 376064 NOL5A 1119139 200880 at 388392 DNAJA1 1130658 200886 s at 447492 PGAM1 1130668 200897 s at 194431 KIAA0992 1130674 200905 x at 381008 HLA-E 1130676 200907 s at 194431 KIAA0992 1130680 200912 s at 511904 EIF4A2 1130687 200924 s at 79748 SLC3A2 1119155 200934 at 110713 DEK 1130704 200951 s at 376071 CCND2 1130707 200956 s at 79162 SSRP1 1130712 200965 s at 442540 ABLIMI 1119171 200974 at 208641 ACTA2 1119173 200978 at 75375 MDH1 1119183 200997 at 211203 RBM4 1130732 201002 s _at 381025 UBE2V1 1119186 201005 at 387579 CD9 1130735 201009 s at 179526 TXNIP 1130744 201027 s at 158688 EIF5B 1130746 201029 s at 283477 CD99 170 1130747 201030 x at 234489 LDHB 1119202 201042 at - 512708 TGM2 1130755 201043 s at 356089 ANP32A 1119209 201063 at 167791 RCN1 1130771 201068 s at 61153 PSMC2 1119212 201069 at 367877 MMP2 1130799 201114 x at 233952 PSMA7 1130812 201131 s at 194657 CDH1 1119237 201141 at 389964 GPNMB 1130820 201144 s at 151777 EIF2S1 1119239 201145 at 199625 HAXI 1130835 201163 s at 435795 IGFBP7 1130839 201167 x at 159161 ARHGDIA 1119243 201171 at 440165 ATP6VOE 1119245 201178 at 5912 FBXO7 1130852 201184 s at 74441 CHD4 1130855 201189 s at 77515 ITPR3 1119251 201194 at 433941 SEPW1 1119258 201209 at 88556 HDAC1 1119260 201212 at 18069 LGMN 1119263 201216 at 511762 C12orf8 1130871 201222 s at 159087 RAD23B 1130879 20.1231 s at 433455 ENOI 1119268 201234 at 6196 ILK 1130882 201236 s at 75462 BTG2 1130888 201244 s at 257266 RAF1 1130898 201260 s at 80919 SYPL 1130900 201262 s at 821 BGN 1130906 201277 s at 81361 HNRPAB 1130910 201284 s at 221589 APEH 1130911 201287 s at 82109 SoC1 1119294 201292 at 156346 TOP2A 1130914 201294 s at 315379 WSB1 1130922 201305 x at 459987 ANP32B 1130923 201306 s at 459987 ANP32B 1130926 201310 s at 508741 C5orfl3 1119300 201314 at 155206 STK25 1130936 201331 s at 437475 STAT6 1130942 201338 x at 445977 GTF3A 1119311 201341 at 104925 ENCI 1119317 201349 at 396783 SLC9A3R1 1119325 201365 at 74563 OAZ2 1119334 201389 at 149609 ITGA5 1130972 201393 s at 76473 IGF2R 1130977 201401 s at 83636 ADRBKI 1119350 201425 at 331141 ALDH2 1130994 201431 s at 150358 DPYSL3 1119361 201448 at 391858 TIA1 1119365 201460 at 75074 MAPKAPK2 1131012 201464 x at 78465 JUN 1119369 201473 at 25292 JUNB 1131019 201474 s at 265829 ITGA3 1119375 201489 at 381072 PPIF 1131038 201502 s at 81328 NFKBIA 171 1119383 201508 at 1516 IGFBP4 1119390 201518 at 77254 CBX1 1119400 201536 at 181046 DUSP3 1119401 201540 at 421383 FHL1 1131068 201564 s at 118400 FSCN1 1131069 201565 s at 180919 ID2 1131074 201572 x at 76894 DCTD 1119417 201579 at 166994 FAT 1131081 201586 s at 180610 SFPQ 1131082 201587 s at 182018 IRAK1 1119424 201599 at 75485 OAT 1131107 201628 s at 432330 RRAGA 1131110 201631 s at 76095 IER3 1119438 201641 at 118110 BST2 1131119 201647 s at 349656 SCARB2 1119443 201648 at 436004 JAK1 1119445 201650 at 309517 KRT19 1119448 201656 at 212296 ITGA6 1131140 201684 s at 194035 C14orf92 1131149 201694 s at 326035 EGR1 1131150 201695 s at 75514 NP 1119460 201696 at 76122 SFRS4 1119462 201700 at 83173 CCND3 1.119466 201710 at 179718 MYBL2 1119467 201714 at 21635 TUBG1 1119475 201739 at 296323 SGK 1119477 201743 at 163867 CD14 1131181 201744 s at 406475 LUM 1119479 201746 at 408312 TPS3 1119488 201761 at 154672 MTHFD2 1131197 201778 s at 269902 KIAA0494 1119503 201803 at 149353 POLR2B 1131218 201809 s at 76753 ENG 1131219 201810 s at 109150 SH3BP5 1119510 201820 at 433845 KRT5 1119515 201833 at 3352 HDAC2 1119516 201834 at 6061 PRKABI 1119519 201849 at 79428 BNIP3 1131246 201853 s at 153752 CDC25B 1131260 201872 s at 12013 ABCE1 1131263 201877 s at 249955 PPP2R5C 1119533 201886 at 283976 WDR23 1131268 201888 s at 285115 IL13RAI 1119537 201895 at 446641 TIMP1 1131274 201897 s at 374378 CKS1B 1119541 201910 at 207428 FARPI 1119546 201921 at 433898 GNG10 1131290 201925 s at 408864 DAF 1119557 201939 at 398157 PLK2 1119559 201941 at 5057 CPD 1119561 201945 at 59242 FURIN 1119564 201952 at ' 10247 ALCAM 1119565 201953 at 135471 CIBi 1119566 201954 at 433506 ARPC1B 172 1119568 201957 at 269777 PPP1R12B 1131321 201983 s at 77432 EGFR 1131325 201990 s at 13313 CREBL2 1119582 201998 at 2554 SIAT1 1131336 202010 s at 405945 ZNF410 1131340 202018 s at 437457 LTF 1131342 202020 s at 13351 LANCL1 1131379 202075 s at 439312 PLTP 1119611 202076 at 289107 BIRC2 1131395 202102 s at 278675 BRD4 1131401 202119 s at 14158 CPNE3 1131405 202123 s at 446504 ABL1 1131407 202125 s at 154248 ALS2CR3 1119633 202126 at 198891 PRPF4B 1119636 202130 at 209061 RIOK3 1131411 202135 s at 2477 ACTR1B 1119639 202136 at 145894 BS69 1131414 202140 s at 511790 CLK3 1119647 202161 at 2499 'PRKCL1 1119652 202175 at 458374 CHPF 1119655 202178 at 407181 1131450 202200 s at 369358 SRPK1 1119667 202206 at 111554 ARL7 1119680 202237 at 364345 NNMT 1119683 202241 at 444947 C8FW 1119684 202242 at 439586 TM4SF2 1131473 202243 s at 89545 PSMB4 1131474 202246 s at 95577 CDK4 1119694 202265 at 380403 COMMD3 1119699 202273 at 307783 PDGFRB 1119706 202281 at 153227 GAK 1119708 202283 at 173594 SERPINF1 1131490 202284 s at 370771 CDKN1A 1119709 202288 at 338207 FRAP1 1131497 202295 s at 114931 CTSH 1131503 202303 x at 135705 SMARCA5 1131507 202311 s at 172928 COLlAl 1119725 202329 at 77793 CSK 1119729 202338 at 164457 TK1 1131531 202350 s at 153647 MATN2 1119734 202351 at 436873 ITGAV 1131541 202369 s at 310230 TRAM2 1119752 202391 at 511745 BASPI 1131561 202403 s at 232115 COL1A2 1119765 202421 at 81234 IGSF3 1119766 202423 at 93231 MYST3 1131578 202431 s at 202453 MYC 1131584 202439 s at 303154 IDS 1131592 202450 s at 83942 CTSK 1131594 202454 s at 306251 ERBB3 1119775 202455 at 9028 HDAC5 1119780 202472 at 75694 MPI 1119782 202478 at 155418 TRB2 1131614 202483 s at 24763 RANBP1 173 1119799 202518 at 408219 BCL7B 1119802 202522 at 7370 PITPNB 1131636 202524 s at 436193 SPOCK2 1131637 202527 s at 75862 MADH4 1119807 202530 at 79107 MAPK14 .1119808 202531 at 80645 IRF1 1131640 202534 x at 83765 DHFR 1131645 202542 s at 105656 SCYE1 1119813 202545 at 155342 PRKCD 1131654 202555 s at 386078 MYLK 1119817 202561 at 409194 TNKS 1131663 202568 s at 437625 MARK3 1119820 202573 at 181390 CSNK1G2 1119826 202589 at 87491 TYMS 1131687 202606 s at 369280 TLK1 1119838 202615 at 469951 GNAQ 1119841 202625 at 80887 LYN 1119846 202634 at 351475 POLR2K 1131705 202638 s at 386467 .ICAMI 1131710 202644 s at 211600 TNFAIP3 1119860 202670 at 132311 MAP2KI 1131733 202686 s at 83341 AXL 1119868 202688 at 387871 TNFSF1O 1131737 202693 s at 9075 STK17A 1119872 202696 at 95220 OSRI 1119873 202697 at 446393 CPSF5 1119876 202703 at 14611 DUSPi1 1119878 202705 at 194698 CCNB2 1119880 202709 at 442844 FMOD 1119884 202716 at 418004 PTPNI 1131752 202724 s at 170133 FOXO1A 1131753 202727 s at 180866 IFNGR1 1131755 202729 s at 241257 LTBPI 1119889 202731 at 257697 PDCD4 1131757 202736 s at 76719 LSM4 1119894 202740 at 334707 ACY1 1119895 202741 at 156324 PRKACB 1119903 202753 at 350939 p44S10 1131767 202758 s at 296776 RFXANK 1119906 202762 at 58617 ROCK2 1119907 202763 at 141125 CASP3 1131778 202779 s at 396393 UBE2S 1119916 202780 at 177584 OXCT 1119919 202786 at 199263 STK39 1119920 202788 at 234521 MAPKAPK3 1119924 202794 at 32309 INPP1 1119928 202799 at 317335 CLPP 1131786 202803 s at 375957 ITGB2 1119936 202811 at 407994 STAMBP 1119939 202820 at 170087 AHR 1119946 202834 at 19383 AGT 1119950 202840 at 402752 TAF15 1131806 202842 s at 6790 DNAJB9 1131808 202845 s at 75447 RALBP1 174 1131813 202853 s at 285346 RYK 1131815 202856 s at 386678 SLC16A3 1131816 202859 x at 624 IL8 1131827 202880 s at 1050 PSCD1 1131835 202888 s at 1239 ANPEP 1119972 202894 at 437008 EPHB4 1131839 202899_s at 405144 SFRS3 1131845 202906 s at 25812 NBS1 1131847 202910 s at 3107 CD97 1119979 202911_at 445052 MSH6 1119983 .202920 at 409783 ANK2 1131854 202923 s at 414985 GCLC 1131861 202933 s at 194148 YES1 1131863 202936 s at 2316 SOX9 1131868 202947 s at 81994 GYPC 1119995 202948 at 82112 IL1RI 1119997 202951 at 367811 STK38 1131870 202952 s at 8850 ADAM12 1119998 202953 at 8986 CIQB 1131875 202965 s at 169172 CAPN6 1120008 202969 at 173135 DYRK2 1120011 202983 at 3068 SMARCA3 1120016 202991 at 77628 STARD3 1120023 203005 at 1116 LTBR 1120024 203006 at 408063 INPP5A 1120026 203010 at 437058 STAT5A 1131916 203035 s at 435761 PIAS3 1131918 203037 s at 77694 MTSSI 1120038 203044 at 110488 CHSY1 1120044 203053 at 22960 BCAS2 1131925 203054 s at 250894 TCTA 1120053 203073 at 82399 COG2 1120055 203075 at 110741 MADH2 1120059 203083 at 458354 THBS2 1131940 203085 s at 1103 TGFB1 1120063 203090 at 118684 SDF2 1120069 203104 at 174142 CSF1R 1120072 203110 at 405474 PTK2B 1131955 203112 s at 21771 WHSC2 1120079 203126 at 5753 IMPA2 1131964 203130 s at 6641 KIF5C 1120081 203131 at 74615 PDGFRA 1120082 203132 at 408528 RBI 1120088 203138 at 13340 HAT1 1120089 203139 at - 244318 DAPK1 1120090 203140 at 155024 BCL6 1131972 203154 s at 20447 PAK4 1131975 203160 s at 24439 RNF8 1120108 203175 at 75082 ARHG 1120120 203196 at 307915 ABCC4 1120121 203198 at 150423 CDK9 1131998 203210 s at 443227 RFC5 1120127 203213 at 334562 CDC2 1132004 203217 s at 415117 SIAT9 175 1120128 203218 at 348446 MAPK9 1120129 203221 at 406491 TLE1 1132011 203229 s at 73986 CLK2 1132013 203232 s at 434961 SCAl 1120134 203233 at 75545 IL4R 1132016 203238 s at 8546 NOTCH3 1120137 203240 at 111732 FCGBP 1132022 203247 s at 173911 ZNF24 1120145 203256 at 191842 CDH3 1132031 203266 s at 134106 MAP2K4 1132034 203271 s at 410455 UNC119 1132035 203272 s at 8186 TUSC2 1120152 203275 at 83795 IRF2 1120153 203276 at 89497 LMNB1 1120160 203288 at 436976 KIAA0355 1120163 203302 at 709 DCK 1132058 203313 s at 161999 TGIF 1120191 203373 at 405946 SOCS2 1120194 203379 at 149957 RPS6KA1 1120196 203386 at 173802 TBC1D4 1132104 203387 s at 173802 TBC1D4 1120205 203405 at 5198 DSCR2 1120214 203416 at 443057 CD53 1120216 203418 at 85137 CCNA2 1132122 203434 s at 307734 MME 1132132 203454 s at 279910 ATOX1 1120254 203485 at 99947 RTNI 1120261 203499 at 171596 EPHA2 1120266 203507 at 246381 CD68 1120267 203508 at 256278 TNFRSF1B 1120269 203510 at 419124 MET 1120272 203514 at 29282 MAP3K3 1120274 203517 at 31584 MTX2 1132159 203521 s at 147868 ZNF318 1120278 203528 at 511748 SEMA4D 1120288 203547 at 17483 CD4 1120289 203552 at 246970 MAP4K5 1132178 203554 x at 350966 PTTG1 1120299 203574 at 79334 NFIL3 1120300 203575 at 82201 CSNK2A2 1132196 203591 s at 381027 CSF3R 1120316 203611 at 63335 TERF2 1120317 203612 at 106880 BYSL 1120324 203627 at 239176 IGF1R 1132220 203632 s at 448805 GPRC5B 1132223 203638 s at 404081 FGFR2 1132230 203649 s at 76422 PLA2G2A 1120335 203652 at 432787 MAP3K11 1132236 203661 s at 374849 TMODI 1120350 203679 at 446686 IL1RL1LG 1120353 203685 at 79241 BCL2 1120355 203687 at 80420 CX3CL1 1120356 203688 at 458291 PKD2 1120359 203697 at 128453 FRZB 176 1132256 203702 s at 169910 KIAA0173 1132260 203706 s at 173859 FZD7 1120361 203708 at 188 PDE4B 1120362 203709 at 196177 PHKG2 1120366 203717 at 44926 DPP4 1120370 203723 at 78877 ITPKB 1120373 203728 at 93213 BAK1 1120378 203738 at 151046 FLJ11193 1120385 203755 at 36708 BUBIB 1120387 203758 at 75262 CTSO 1120389 203761 at 75367 SLA 1132288 203767 s at 79876 STS 1132292 203771 s at 435726 BLVRA 1132294 203777 s at 32156 RPS6KB2 1120400 203787 at 152207 SSBP2 1120402 203794 at 18586 CDC42BPA 1132306 203795 s at 371758 BCL7A 1120417 203827 at 9398 FLJ10055 1120419 203830 at 9800 NJMU-R1 1120422 203835 at 151641 GARP 1120423 203837 at 151988 MAP3K5 1132329 203839 s at 528296 ACK1 1120425 203843 at 188361 RPS6KA3 1132336 203853 s at 30687 GAB2 1120433 203856 at 422662 VRK1 1132345 203868 s at 109225- VCAM1 1120438 203870 at 109268 USP46 1132349 203881 s at 169470 DMD 1132353 203887 s at 2030 THBD 1132354 203890 s at 153908 DAPK3 1120465 203915 at 77367 CXCL9 1120477 203934 at 12337 KDR 1120478 203935 at 150402 ACVR1 1132375 203942 s at 157199 MARK2 1132376 203944 x at 169963 BTN2A1 1120483 203947 at 180034 CSTF3 1120484 203949 at 458272 MPO 1120494 203967 at 405958 CDC6 1120500 203979 at 82568 CYP27A1 1132396 203988 s at 118722 FUT8 1120509 204000 at 155090 GNB5 1132407 204005 s at 406074 PAWR 1120520 204023 at 35120 RFC4 1120524 204033 at 436187 TRIP13 1120529 204039 at 76171 CEBPA 1132426 204049 s at 102471 C6orf56 1132428 204051 s at 105700 SFRP4 1120538 204057 at 14453 ICSBP1 1132433 204059 s at 14732 ME1 1132434 204060 s at 147996 PRKX 1132435 204062 s at 168762 ULK2 1120544 204068 at 166684 STK3 1120553 204086 at 30743 PRAME 1120555 204090 at 444 STK19 177 1132449 204092 s at 250822 STK6 1120562 204103 at 75703 CCL4 1120564 204106 at 79358 TESKI 1120572 204116 at 84 IL2RG 1120574 204118 at 901 CD48 1132460 204126 s at 114311 CDC45L 1120580 204127 at 115474 RFC3 1120581 204129 at 415209 BCL9 1132462 204131 s at 14845 FOXO3A 1120583 204133 at 153768 RNU3P2 1120588 204140 at 421194 TPST1 1132468 204147 s at 79353 TFDP1 1120593 204150 at 301989 STABI 1120594 204154 at 442378 CDOl 1120595 204156 at 444909 KIAA0999 1120596 204159 at 4854 CDKN2C 1120601 204166 at 441129 KIAA0963 1132479 204170 s at 83758 CKS2 1120605 204171 at 86858 RPS6KB1 1132485 204183 s at 445563 ADRBK2 1120615 204191 at 181315 IFNAR1 1120616 204192 at 166556 CD37 1120617 204193 at 439777 CPT1B 1120625 204208 at 27345 RNGTT 1132498 204211 x at 439523 PRKR 1120630 204218_at 38044 DKFZP564M 082 1132504 204222 sat 511765 GLIPR1 1120633 204225 at 222874 HDAC4 1120637 204232 at 433300 FCERIG 1132519 204247 s at 166071 CDK5 1132520 204249 s at 283063 LMO2 1120643 204252 at 19192 CDK2 1132525 204255 s at 2062 VDR 1120645 204257 at 21765 FADS3 1132529 204265 s at 288316 GPSM3 1132531 204267 x at 77783 PKMYT1 1120651 204269 at 80205 PIM2 1132536 204285 s at 96 PMAIP1 1120673 204301 at 5333 KIAA0711 1132545 204306 s at 512857 CD151 1132547 204310 s at 78518 NPR2 1120695 204352 at 385685 TRAF5 1120697 204355 at 323462 DHX30 1132572 204357 s at 36566 LIMKI 1120700 204362 at 410745 SCAP2 1120703 204368 at 83974 SLCO2A1 1132584 204379 s at 1420 FGFR3 1120716 204392 at 512804 CAMK1 1120717 204394 at 444159 SLC43AI 1132592 204396 s at 211569 GRK5 1120720 204401 at 10082 KCNN4 1120730 204415 at 287721 G1P3 1120743 204440 at 79197 CD83 178 1132614 204446 s at 89499 ALOX5 1120750 204454 at 45231 LDOC1 1132628 204468 s at 78824 TIE 1120755 204470 at 789 CXCL1 1120765 204484_at 343329 PIK3C2B 1132636 204490 s at 306278 CD44 1120770 204493 at 300825 BID 1120779 204510 at 28853 CDC7 1120780 204511 at 301283 FARP2 1120785 204517_at 110364 PPIC 1120789 204524 at 154729 PDPK1 1132651 204529 s at 439767 TOX 1120792 204533 at 413924 CXCLIO 1120803 204549 at 321045 IKBKE 1120808 204562 at 127686 IRF4 1120809 204563 at 82848 SELL 1120813 204568 at 414809 KIAA0831 1120814 204569 at 417022 ICK 1120818 204579_ at 165950 FGFR4 1120824 204589 at 200598 ARK5 1120825 204591 at 388344 CHL1 1120828 204600_at 2913 EPHB3 1120832 204604 at 57856 PFTK1 1120834 204606 at 57907 CCL21 1120838 204612_at 433700 PKIA 1120839 204613 at 512298 PLCG2 1120846 204632 at 105584 RPS6KA4 1132700 204633 s at 109058 RPS6KA5 1120847 204634 at 433008 NEK4 1120853 204641 at 153704 NEK2 1120854 204642 at 154210 EDGI 1120858 204647 at 410683 HOMER3 1120863 204655 at 489044 CCL5 1120875 204674 at 124922 LRMP 1120880 204682 at 105689 LTBP2 1120881 204683_at 433303 ICAM2 1132726 204707 s at 433728 MAPK4 1120900 204718 at 380089 EPHB6 1132734 204724 s at 126248 COL9A3 1120918 204754 at 250692 HLF 1120923 204765 at 334 ARHGEF5 1120925 204773 at 204891 IL1IRA 1132762 204777 s at 80395 MAL 1132766 204781 s at 82359 TNFRSF6 1132768 204785 x at 512211 IFNAR2 1132775 204803 s at 1027 RRAD 1132780 204811 s at 389415 CACNA2D2 1120946 204813 at 25209 MAPK1O 1120952 204822 at 169840 TTK 1120955 204825 at 184339 MELK 1120958 204831 at 397734 CDK8 1132787 204832 s at 2534 BMPR1A 1132799 204859 s at 373575 APAFi 1120976 204867 at 245644 GCHFR 179 1120980 204872 at 494269 TLE4 1132809 204878 s at 291623 PSK 1120986 204886 at 172052 PLK4 1132818 204891 s at 1765 LCK 1132825 204900 x at 512813 SAP30 1132830 204908 s at 31210 BCL3 1120993 204912 at 327 IL1ORA 1132834 204914 s at 432638 SOXi1 1121000 204924 at 519033 TLR2 1121005 204932 at 81791 TNFRSF11B 1121007 204936 at 440835 SF1 1121012 204947 at 96055 E2FI 1121013 204949 at 353214 ICAM3 1132850 204954 s at 130988 DYRK1B 1121021 204958 at 153640 PLK3 1132851 204961 s at 1583 1132852 204962 s at 1594 CENPA 1121028 204968 at 247323 APOM 1121029 204971 at 412999 CSTA 1121033 204975 at 356835 EMP2 1132860 204986 s at 291623 PSK 1132862 204990 s at 85266 ITGB4 1132866 204998 s at 9754 ATF5 1132874 205013 s at 197029 ADORA2A 1121054 205016 at 170009 TGFA 1121057 205026 at 434992 STAT5B 1132883 205027 s at 432453 MAP3K8 1121061 205032 at 387725 ITGA2 1121062 205034 at 408658 CCNE2 1132890 205049 s at 79630 CD79A 1132892 205051 s at 81665 KIT 1121073 205052 at 81886 AUH 1121076 205055 at 389133 ITGAE 1121082 205067 at 126256 IL1B 1121100 205098 at 301921 CCR1 1121102 205101 at 126714 MHC2TA 1132918 205114 s at 73817 CCL3 1132920 205119 s at 753 FPR1 1121115 205124 at 78881 MEF2B 1121117 205126 at 82771 VRK2 1121120 205130 at 104119 RAGE 1121129 205159 at 285401 CSF2RB 1121136 205168 at 440905 DDR2 1132953 205180 s at 86947 ADAM8 1121143 205184 at 447973 GNG4 1121149 205192 at 440315 MAP3KI4 1132959 205198 s at 606 ATP7A 1121159 205205 at 307905 RELB 1121161 205207 at 512234 IL6 1132961 205212 s at 337242 CENTBI 1121166 205214 at 88297 STK17B 1121170 205220 at 458425 HM74 1121186 205242 at 100431 CXCL13 1121190 205247 at 436100 NOTCH4 180 1121195 205253 at 408222 PBXi 1132973 205255 x at 169294 TCF7 1121201 205263 at 193516 BCL10 1121203 205266 at 2250 LIF 1121205 205269 at 2488 LCP2 1132979 205271_ s at 26322 CCRK 1121217 205291 at 75596 IL2RB 1121220 205296 at 87 RBLI 1132990 205297 s at 89575 CD79B 1132994 205301 s at 380271 OGG1 1132996 205306 x at 409081 KMO 1121228 205312 at 157441 SPl1 1133004 205327 s at 389846 ACVR2 1121248 205345 at 54089 BARD1 1133011 205347 s at 56145 TMSNB 1121265 205372 at 14968 PLAGI 1133021 205377 s at 154495 ACHE 1133024 205383 s at 436987 ZNF288 1133030 205392 s at 272493 CCLI5 1121276 205394 at 24529 CHEKI 1121278 205399 at 21355 DCAMKL1 1121281 205403 at 25333 IL1R2 1121287 205411 at 35140 STK4 1121290 205418 at 7636 FES 1121291 205419 at 784 EB12 1133042 205422 s at 311054 ITGBL1 1133047 205434 s at 528338 AAKI 1133049 205436 s at 147097 H2AFX 1121301 205437 at 449971 ZNF134 1121306 205443 at 179312 SNAPC1 1121309 205449 at 23642 HSU79266 1121315 205455 at 2942 MST1R 1121316 205456 at 3003 CD3E 1121322 205467 at 5353 CASPIO 1121326 205476 at 75498 CCL20 1133065 205479 s at 77274 PLAU 1133068 205483 s at 458485 G1P2 1121329 205484 at 88012 SIT 1121331 205486 at 8980 TESK2 1121343 205504 at 159494 BTK 1133076 205512 sat 18720 PDCD8 1133080 205526 s at 440341 KATNAl 1133091 205544 s at 73792 CR2 1133093 205546 s at 75516 TYK2 1121368 205551_at 8071 SV2B 1133099 205554 s at 88646 DNASE1L3 1121371 205558 at 444172 TRAF6 1133102 205565 s at 360041 FRDA 1121380 205569 at 10887 LAMP3 1121383 205572 at 115181 ANGPT2 1121387 205578 at 208080 ROR2 1133111 205593 s at 389777 PDE9A 1121400 205599 at 223474 TRAF1 1133117 205607 s at 435560 PACE-1 18 1 1121404 205609 at 2463 ANGPT1 1121406 205611 at TNFSF12 1121408 205613 at 258326 LOC51760 1133119 205614 x at 512587 MSTI 1121414 205621 at 94542 ALKBH 1121436 205659 at 487662 HDAC9 1121444 205668 at 153563 LY75 1133138 205671 s at 1802 HLA-DOB 1133141 205677 s at 344524 DLEU1 1121452 205681 at 227817 BCL2A1 1133148 205692 s at 174944 CD38 1133150 205698 s at 256924 MAP2K6 1121468 205707 at 129751 IL17R 1133156 205713 s at 1584 COMP 1121473 205718 at 1741 ITGB7 1121482 205729 at 238648 OSMR 1121497 205758 at 85258 CD8A 1121511 205780 at 155419 BIK 1133184 205786 s at 172631 ITGAM 1121516 205789 at '1799 CD1D 1121518 205792 at 194679 WISP2 1133192 205801 s at 24024 RASGRP3 1133195 205805 sat 274243 RORI 1121533 205821 at 387787 KLRK1 1121542 205831 at 89476 CD2 1133210 205842 s at 434374 JAK2 1121546 205844 at 12114 VNN1 1121554 205854 at 437046 TULP3 1121558 205858 at 415768 NGFR 1121559 205859_at 184018 LY86 1121560 205861 at 437905 SPIB 1121564 205865 at 437783 ARID3A 1133216 205872 x at 502577 PDE4DIP 1121572 205876 at 446501 LIFR 1121573 205878_at 2815 POU6F1 1133219 205879 x at 350321 RET 1121574 205880 at 2891 PRKCM 1133227 205895 s_at 75337 NOLCI 1121584 205898 at 78913 CX3CR1 1121585 205899 at 417050 CCNAI 1121587 205901 at 371809 PNOC 1121589 205904 at 90598 MICA 1133232 205910 s at 406160 CEL 1121629 205965 at 41691 BATF 1133252 205977 s at 89839 EPHAl 1121643 205986 at 514575 AATK 1121645 205988 at 398093 CD84 1133260 205992 s at 528402 ILl5 1121650 206002 at 421137 GPR64 1121655 206009 at 222 ITGA9 1133272 206028 s at 306178 MERTK 1133275 206036 s at 44313 REL 1121680 206049 at 73800 SELP 1133296 206070 s at 123642 EPHA3 182 1133299 206075 s at 446484 CSNK2A1 1121689 .206076 at 155586 B7 1121693 206080 at 170156 KIAA0450 1121695 206082 at 511759 HCP5 1133300 206085 s at 19904 CTH 1121711 206106 at 432642 MAPK12 1121717 206114 at 73964 EPHA4 1121720 206118 at 80642 STAT4 1121722 206120 at 83731 CD33 1121726 206126 at 113916 BLR1 1121739 206142 at 85863 ZNF135 1121743 206148 at 460433 IL3RA 1121745 206150 at 355307 TNFRSF7 1121757 206170 at 2551 ADRB2 1121759 206172 at 336046 IL13RA2 1121760 206176 at 285671 BMP6 1121762 206181 at 32970 SLAMF1 1121767 206187 at 458324 PTGIR 1121780 206206 at 87205 LY64 1121783 206211 at 89546 SELE 1121788 206216 at 104865 STK23 1121792 206222 at 119684 TNFRSF10C 1121793 206223 at 122708 LMTK2 1121809 206247 at 211580 MICB 1121814 206255 at 389900 BLK 1133355 206267 s at 437808 MATK 1121828 206271 at 29499 TLR3 1121834 206279 at 183165 PRKY 1133358 206283 s at 73828 TAL1 1121841 206291 at 80962 NTS 1121844 206295 at 83077 IL18 1121848 206301 at 278005 TEC 1121853 206310 at 98243 SPINK2 1121854 206312 at 171470 GUCY2C 1121857 206315 at 114948 CRLF1 1133376 206324 s at 129208 DAPK2 1121869 206336 at 164021 CXCL6 1121870 206337 at 1652 CCR7 1121874 206341 at 130058 IL2RA 1133388 206362 x at 435014 MAP3K10 1121887 206363 at 134859 MAF 1133389 206366 x at 174228 XCL1 1133392 206380 s at 53155 PFC 1133397 206390 x at 81564 PF4 1133400 206398 s at 96023 CD19 1133405 206407 s at 414629 CCL13 1133406 206411 s at 159472 ABL2 1121918 206412 at 121558 FER 1133407 206413 s at 144519 TCL6 1133408 206414 s at 12802 DDEF2 1121947 206464 at 27372 BMX 1133430 206467 x at 348183 TNFRSF6B 1121953 206478 at 38365 KIAA0125 1121956 206482 at 51133 PTK6 183 1121959 206486 at 409523 LAG3 1121963 206493 at 411312 ITGA2B 1121966 206498 at 82027 OCA2 1133445 206499 s at 196769 CHCI 1121970 206508 at 99899 TNFSF7 1133453 206518 s at 117149 RGS9 1121996 206545 at 1987 CD28 1122007 206569 at 411311 IL24 1133476 206571 s at 3628 MAP4K4 1122009 206575 at 50905 CDKL5 1122021 206591 at 73958 RAG1 1122036 206618 at 159301 IL18R1 1122051 206637 at 2465 GPR105 1122053 206641 at 2556 TNFRSF17 1122065 206660 at 348935 IGLL1 1122075 206674 at 385 FLT3 1133515 206687 s_ at 63489 PTPN6 1122087 206693 at 72927 IL7 1122091 206702 at 89640 TEK 1122104 206718 at 1149 LMOI 1122112 206729 at 1314 TNFRSF8 1122131 206756 at 138155 CHST7 1133538 206760 s at 1416 FCER2 1122139 206766 at 158237 ITGA10 1122156 206794 at 1939 ERBB4 1122165 206804 at 2259 CD3G 1122181 206828 at 29877 TXK 1133565 206846 sat 6764 HDAC6 1133568 206854 s at 290346 MAP3K7 1133569 206855 s at 76873 HYAL2 1133576 206864 s at 87247 HRK 1133577 206874 s at 105751 SLK 1133580 206881 s at 113277 LILRA3 1122215 206887 at 528317 CCBP2 1122217 206890 at 223894 IL12RB1 1122219 206892 at 437877 AMHR2 1122230 206907 at 1524 TNFSF9 1122241 206923 at 349611 PRKCA 1133595 206926 s at 1721 ILI1 1122253 206943 at 28005 TGFBR1 1122274 206974 at 34526 CXCR6 1122275 206975 at 36 LTA 1122277 206978 at 511794 CCR2 1122281 206983 at 46468 CCR6 1122284 206988 at 310511 CCL25 1133618 206991 s at 511796 CCR5 1122288 206999 at 413608 IL12RB2 1122292 207008 at 846 IL8RB 1133629 207011 s at 90572 PTK7 1122304 207029 at 1048 KITLG 1122327 207061 at 137575 ERNI 1122335 207073 at 143241 CDKL2 1133652 207076 s at 160786 ASS 1122344 207094 at 194778 IL8RA 1 84 1122353 207111 at 2375 EMR1 1133672 207113 s at 241570 TNF 1133676 207121 _s at 271980 MAPK6 1122380 207160 at 673 IL12A 1133694 207163 s at 368861 AKTI 1122382 207165 at 72550 HMMR 1133700 207173 x at 443435 CDH11 1133701 207176 s at 838 CD80 1133702 207178 s at 89426 FRK 1122388 207179 at 89583 TLX1 1133704 207181 s at 9216 CASP7 1122394 207188 at 100009 CDK3 1133708 207194 s at 512159 ICAM4 1122400 207199 at 439911 TERT 1122412 207216 at 177136 TNFSF8 1122420 207228 at 158029 PRKACG 1133724 207239 s at 171834 PCTK1 1122428 207245 at 183596 UGT2BI7 1133731 207253 s at 21479 UBN1 1122449 207277 at 278694 CD209 1122471 207312 at 512612 PHKG1 1133753 207314 x at 380156 KIR3DL1 1133755 207318 s at 404501 CDC2L5 1133757 207320 x at 6113 STAU 1133766 207339 s at 376208 LTB 1122491 207354 at 10458 CCL16 1133778 207375 s at 12503 ILl 5RA 1133786 207396 s at 153591 ALG3 1133801 207426 s at 181097 TNFSF4 1133802 207428 x at 454861 CDC2L2 1122537 207433 at 193717 IL10 1122541 207442 at 2233 CSF3 1133810 207445 s at 225946 CCR9 1122544 207446 at 366986 TLR6 1133829 207497_ s at 386748 MS4A2 1122581 207505 at 41749 PRKG2 1133834 207509 s at 43803 LAIR2 1122596 207533 at 72918 CCLI 1133846 207536 s at 528403 TNFRSF9 1122599 207538 at 73917 IL4 1133848 207540 s at 192182 SYK 1122602 207550 at 82906 CDC20 1122609 207568 at 103128 CHRNA6 1122610 207569 at 1041 ROSI 1133867 207571 x at 10649 Clorf38 1133869 207574 s at 110571 GADD45B 1133901 207633 s at 156465 MUSK 1122640 207634 at 158297 PDCD1 1122645 207641 at 158341 TNFRSF13B 1133904 207643 s at 159 TNFRSFIA 1133910 207655 s at 167746 BLNK 1122664 207681 at 198252 CXCR3 1133931 207697 x at 306230 LILRB2 1122680 207709 at 256067 PRKAA2 185 1122710 207766 at 380788 CDKLI 1133998 207826 s at 76884 ID3 1122738 207840 at 81743 CD160 1122740 207844 at 845 IL13 1122743 207849 at 89679 IL2 1122744 207850 at 89690 CXCL3 1122749 207861 at 80420 CX3CL1 1122763 207884 at 309958 GUCY2D 1122767 207892 at 652 TNFSF5 1122772 207900 at 66742 CCL17 1122773 207901 at 674 IL12B 1122774 207902 at 68876 ILRA 1122775 207906 at 694 IL3 1122776 207907 at 129708 TNFSF14 1122796 207952 at 2247 IL5 1134069 207979 s at 405667 CD8B1 1134076 207988 s at 83583 ARPC2 1134083 207996 s at 285091 C18orfd 1134095 208018 s at 89555 , HCK 1134109 208037 s at 102598 MADCAMI 1122824 208038 at 416814 ILIRL2 1122834 208059 at 113222 CCR8 1134133 208075 s at 251526 CCL7 1134145 208091_sat 4750 DKFZP564K 0822 1134200 208161 s at 90786 ABCC3 1134212 208178 x at 367689 TRIO 1134220 208189 s at 370421 MYO7A 1122863 208193 at 960 IL9 1122864 208195 at 434384 TTN 1122865 208200 at 1722 IL1A 1134230 208206 s at 99491 RASGRP2 1134233 208212 s at 410680 ALK 1134270 208284 x at 352119 GGT1 1134271 208286 x at 249184 POU5F1 1134280 208303 s at 287729 CRLF2 1122914 208304 at 506190 CCR3 1134296 208335 s at 183 FY 1134316 208365 s at 32959 GRK4 1122939 208376 at 184926 CCR4 1134361 208426 x at 515605 KIR2DL4 1134370 208438 s at 1422 FGR 1122956 208450 at 113987 LGALS2 1134379 208451 s at 150833 C4A 1122983 208495 at 249125 TLX3 1122994 208524 at 159900 GPR15 1134422 208536 s at 84063 BCL2L1I 1134424 208540 x at S100A14 1123026 208578 at 250443 SCNIOA 1134457 208605 s at 406293 NTRK1 1134480 208634 s at 372463 MACF1 1123038 208636 at 119000 ACTN1 1123039 208638 at 212102 P5 1134494 208657 s at 288094 MSF 186 1123052 208680 at 180909 PRDX1 1123053 208683 at 350899 CAPN2 1134517 208690 s at 75807 PDLIM1 1123055 208691 at 185726 TFRC 1134523 208700 s at 89643 TKT 1134532 208711 s at 371468 CCND1 1134533 208716 s at 93832 LOC54499 1134542 208729 x at 77961 HLA-B 1123086 208774 at 378918 CSNK1D 1134582 208794 s at 78202 SMARCA4 1134593 208812 x at 274485 HLA-C 1123105 208820 at 434281 PTK2 1123108 208828 at 108112 POLE3 1134615 208851 s at 134643 THY1 1134618 208854 s at 168913 STK24 1134647 208892 s at 298654 DUSP6 1123127 208894 at 409805 HLA-DRA 1134653 208901 s at 253536 TOP1 1134665 208921 s at 422340 SRI 1134674 208937 s at 410900 IDI 1134676 208942 s at 158193 TLOCI 1123148 208944 at 82028 TGFBR2 1134679 208946 s at 12272 BECN1 1134682 208949 s at 411701 LGALS3 1134687 208959 s at 154023 TXNDC4 1134699 208974 x at 439683 KPNB1 1123160 208982 at 78146 PECAM1 1134706 208987 s at 219614 FBXL1I 1123163 208991 at 421342 STAT3 1134710 208993 s at 77965 PPIG 1134727 209018 s at 439600 PINK1 1134738 . 209033 s at 75842 DYRK1A 1134753 209053 s at 110457 WHSC1 1134778 .209085 x at 166563 RFC1 1123188 209089 at 73957 RAB5A 1123192 209100 at 315177 IFRD2 1123193 209101 at 410037 CTGF 1123198 209112 at 238990 CDKNIB 1134797 209118 s at 433394 TUBA3 1123213 209154 at 12956 TIP-1 1123223 209173 at 226391 AGR2 1134837 209185 s at 143648 IRS2 1123231 209189 at 25647 FOS 1123233 209193 at 81170 PIM1 1123235 209196 at 436930 C6orf 11 1134843 209199 s at 368950 MEF2C 1134850 209210 s at 270411 PLEKHC1 1134852 209214 s at 374477 EWSR1 1134858 209226 s at 405954 TNPO1 1123250 209233 at 135643 C2F 1123255 209239 at 160557 NFKB1 1134865 209241 x at 112028 MINK 1134880 209265 s at 168799 METTL3 1134888 209278 s at 438231 TFP12 187 1123278 209295 at 51233 TNFRSF10B 1134903 209306 s at 153026 SWAP70 1123286 209311 at 410026 BCL2L2 1123289 209317 at 5409 POLR1C 1123293 209333 at 47061 ULK1 1123298 209339 at 20191 SIAH2 1134921 209341 s at 413513 IKBKB 1134928 209352 s at 13999 SIN3B 1123304 209354 at 279899 TNFRSF14 1134933 209360 s at 410774 RUNX1 1123308 209364 at 76366 BAD 1123310 209368 at 212088 EPHX2 1134945 209379 s at 81897 KIAA1 128 1123317 209386 at 351316 TM4SF1 1123321 209392 at 23719 ENPP2 1123331 209409 at 512118 GRB10 1134961 209417 s at 50842 IF135 1123346 209443at 76353 SERPINA3 1123358 209464 at 442658 AURKB 1134988 209467 s at 79516 MKNK1 1134991 209474 s at 444105 ENTPD1 1123369 209481 at 79025 SNRK 1123372 209487 at 195825 RBPMS 1123376 209496 at 37682 RARRES2 1135002 209500 x at 54673 TNFSF13 1123399 209541_at 308053 IGF1 1135023 209543 s at 374990 CD34 1135024 209545 s at 103755 RIPK2 1123401 209550 at 50130 NDN 1135028 209555 s at 443120 CD36 1123413 209575_at 418291 ILIORB 1135042 209582 s at 79015 MOX2 1135047 209589 s at 125124 EPH82 1123419 209590 at 170195 BMP7 1135056 209604 s at 169946 GATA3 1123429 209619 at 446471 CD74 1135068 209621 sat 71719 PDLIM3 1123430 209622 at 153003 STK16 1123437 209636 at 73090 NFKB2 1123439 209642_at 287472 BUBI 1135080 209644 x at 421349 CDKN2A 1135085 209650 s at 505862 C22orf4 1135088 209656 s at 8769 TM4SF1O 1135093 209666 s at 198998 CHUK 1135101 209680 s at 20830 KIFC1 1123455 209682 at 436986 CBLB 1123457 209684 at 446304 RIN2 1135102 209685 s at 349845 PRKCBI 1123459 209687 at 436042 CXCL12 1123470 209704 at 31016 M96 1123476 209711. at 82635 SLC35DI 1123479 209716 at 173894 CSF1 1123490 209732 at 85201 CLECSF2 1123497 209747 at 2025 TGFB3 188 1135130 209757 s at 25960 MYCN 1123502 209760 at 511944 KIAA0922 1123507 209770 at 284283 BTN3A1 1135138 209771 x at 375108 CD24 1135141 209774 x at 75765 CXCL2 1135151 209790 s at 3280 CASP6 1123529 209815 at 159526 PTCH 1135164 209825 s at 458360 UMPK 1135165 209827 s at 170359 ILl6 1123535 209829_at 389488 C6orf32 1135168 209831 x at 118243 DNASE2 1135173 209841 s at 3781 LRRN3 1135186 209860 s at 386741 ANXA7 1135189 209863 s at 137569 TP73L 1123552 209879 at 423077 SELPLG 1135209 209899 s at 74562 SIAHBP1 1123566 209906 at 155935 C3AR1 1135214 209908 s at 169300 TGFB2 1123573 209924 at 16530 CCLI8 1135226 209929 s at 43505 IKBKG 1135227 209930 s at 75643 NFE2 1135229 209932 s at 367676 DUT 1123581 209941 at 390758 RIPK1 1135234 209945 s at 282359 GSK3B 1123584 209946 at 79141 VEGFC 1123586 209948 at 93841 KCNMB1 1123587 209949 at 949 NCF2 1135240 209955 s at 436852 FAP 1135251 209969 s at 21486 STAT1 1135253 209971 x at 301613 JTV1 1135267 209995 s at 2484 TCL1A 1135270 209999 x at 50640 SOCS1 1123608 210017 at 180566 MALT1 1135285 210024 s at 449501 UBE2E3 1123611 210029 at 840 INDO 1123613 210031 at 97087 CD3Z 1123614 210038 at 408049 PRKCQ 1135299 210044 s at 46446 LYL1 1123622 210051 at 8578 EPAC 1123628 210058 at 178695 MAPKI3 1123634 210072 at 50002 CCL19 1123635 210073 at 408614 SIAT8A 1123643 210092 at 421576 MAGOH 1135322 210095 s at 450230 IGFBP3 1135328 210105 s at 390567 FYN 1123663 210133_at 54460 CCL1I 1123671 210148 at 30148 HIPK3 1135350 210151 s at 164267 DYRK3 1123672 210152_at 67846 LILRB4 1123679 210163 at 103982 CXCL11 1123680 210164 at 1051 GZMB 1123682 210166 at 114408 TLR5 1123690 210176 at 111805 TLR1 1123694 210184 at 385521 ITGAX 189 1135374 210214 s at 53250 BMPR2 1135379 210225 x at 511766 LILRB3 1135380 210229 s at 1349 CSF2 1135383 210240 s at 435051 CDKN2D 1123731 210258 at 17165 RGS13 1135395 210260 s at 17839 TNFAIP8 1135399 210275 s at 406096 ZNF216 1123744 210279 at 88269 GPR18 1123760 210313 at 406708 ILT7 1123762 210316 at 415048 FLT4 1123778 210349 at 440638 CAMK4 1123780 210354 at 856 IFNG 1135467 210404 x at 321572 CAMK2B 1135475 210416 s at 146329 CHEK2 1135487 210432 s at 300717 SCN3A 1135489 210438 x at 288178 SSA2 1123814 210439 at 56247 ICOS 1123816 210442_at 66 IL1RL1 1135492 210448 s at 408615 P2RX5 1135513 210481 s at 421437 CD209L 1123833 210487 at 397294 DNT 1123842 210506 at 457 FUT7 1135526 210512 s at 73793 VEGF 1135529 210517 s at 197081 AKAP12 1123847 210523 at 87223 BMPR1B 1135541 210538 s at 127799 BIRC3 1135549 210549 s at 169191 CCL23 1135550 210550 s at 221811 RASGRF1 1135571 210582 s at 278027 LIMK2 1135583 210606 x at 41682 KLRD1 1123875 210607 at 428 FLT3LG 1135592 210621 s at 758 RASAl 1135593 210622_x at 77313 CDK10 1123889 210643 at 333791 TNFSF11 1135606 210644 s at 407964 LAIR1 1123890 210654 at 129844 TNFRSFIOD 1123892 210659 at 159553 CMKLR1 1135622 210671 x at 445864 MAPK8 1135645 210715 s at 31439 SPINT2 1135665 210749 x at 423573 DDR1 1135673 210759 s at 82159 PSMA1 1123938 210772 at 99855 FPRL1 1135684 210775 x at 329502 CASP9 1135685 210776 x at 371282 TCF3 1135735 210838 s at 410104 ACVRL1 1135743 210847 x at 299558 TNFRSF25 1123954 210865 at 2007 TNFSF6 1135755 210869 s at 511397 MCAM 1135773 210889 s at 126384 FCGR2B 1135778 210895 s at 27954 CD86 1135795 210933 s at 55923 Lin1O 1135801 210943 s at 130188 CHS1 1135802 210944 s at 439343 CAPN3 1135826 210976 s at 75160 PFKM 190 1135830 210981 s at 235116 GRK6 1135835 210986 s at 133892 TPM1 1123988 211005 at 498997 LAT 1135852 211008 s at 302903 UBE21 1135858 211015 s at 90093 HSPA4 1135866 211026 s at 409826 MGLL 1135871 211031 s at 104717 CYLN2 1135899 211070 x at 78888 DBI 1135925 211100 x at 149924 LILRB1 1135929 211105 s at 96149 NFATC1 1135930 211107 s at 98338 AURKC 1135966 211155 s at 1166 THPO 1135968 211160 x at 119000 ACTN1 1135974 211168 s at 388125 RENT1 1135982 211178 s at 129758 PSTPIP1 1135994 211197_s at 14155 ICOSL 1136002 211208 s at 288196 CASK 1124049 211276 at 401835 my048 1136048 211282 x at 299558 TNFRSF25 1136051 211286 x at 520937 CSF2RA 1136055 211296 x at 183704 UBC 1136056 211297 s at 184298 CDK7 1136087 211339 s at 211576 ITK 1136109 211370 s at 436145 MAP2K5 1136150 211432 s at 381282 TYRO3 1136152 211434 s at 458436 CCRL2 1136162 211453 s at 326445 AKT2 1136172 211470 s at 38084 SULT1C1 1136185 211488 s at 355722 TGB8 1136193 211499 s at 57732 MAPKI11 1136216 211528 x at 512152 HLA-G 1136269 211593 s at 101474 MAST2 1136273 211597 s at 13775 HOP 1136285 211615 s at 182490 LRPPRC 1124132 211658 at 432121 PRDX2 1136329 211675 s at 132739 HIC 1136337 211685 s at 90063 NCALD 1136343 211692 s at 87246 BBC3 1124137 211693 at 366 MGC27165 1136357 211709 s at 512680 SCGF 1136362 211714 x at 356729 OK/SW-c.56 1136369 211724 x at 387140 FLJ20323 1136371 211726 s at 361155 FMO2 1136379 211734 s at 897 FCERlA 1136391 211748 x at 446429 PTGDS 1136393 211750 x at 406578 TUBA6 1136401 211761 s at 27258 SIP 1136408 211771 s at 1101 POU2F2 1136427 211795 s at 276506 FYB 1136430 211798 x at 102950 IGLJ3 1136459 211828 s at 252550 KIAA0551 1136464 211833 s at 159428 BAX 1136540 211924 s at 179657 PLAUR 1124176 211966 at 407912 COL4A2 192 1124177 211967 at 172089 PORIMIN 1124178 211969_at 446579 HSPCA 1124187 211986 at 378738 MGC5395 1124188 211987 at 282346 TOP2B 1136573 211991 s at 914 HLA-DPAI 1124192 211992 at 275999 PRKWNK1 1124195 211998 at 180877 H3F3B 1136585 212022 s at 80976 MK167 1124215 212037 at 409965 PNN 1136595 212038 s at 404814 VDAC1 1136599 212046 x at 861 MAPK3 1136601 212048 s at 322735 YARS 1136605 212064 x at 448398 MAZ 1124237 212080 at 258855 MLL 1136620 212091 s at 415997 COL6A1 1124254 212110 at 301743 SLC39A14 1124266 212123_at 438991 DKFZP564D 116 1124283 212144 at 406612 UNC84B 1124296 212158 at 1501 SDC2 1124304 212168 at 166887 CPNE1 1124316 212186 at 449863 ACACA 1124318 212190 at 21858 SERPINE2 1124321 212196 at 529772 1136655 212218 s at 388387 FBXO9 1124342 212230 at 432840 PPAP2B 1136662 212240 s at 6241 PIK3R1 1124357 212247 at 413636 NUP205 1124362 212252 at 297343 CAMKK2 1124365 212261 at 334871 TNRC15 1124377 212282 at 199695 MAC30 1124381 212288 at 440808 FNBPI 1124384 212291 at 12259 HIPK1 1124391 212299 at 7200 NEK9 1136681. 212303 x at 91142 KHSRP 1124400 212312 at 305890 BCL2L1 1124411 212326 at 194737 VPS13D 1124416 212331 at 283604 RBL2 1124429 212344 at 409602 SULF1 1136687 212345 s at 59943 CREB3L2 1124438 212358 at 7357 CLIPR-59 1136692 212359 s at 65135 KIAA0913 1124456 212382 at 359289 TCF4 1136702 212399 s at 155584 KIAA0121 1136710 212429 s at 75782 GTF3C2 1136712 212442 s at 503941 LOC253782 1136718 212459 x at 446476 SUCLG2 1136722 212481 s at 250641 TPM4 1136724 212491 s at 433540 DNAJC8 1124539 212494 at 6147 TENC1 1124543 212500 at 99821 ClOorf22 1124549 212508 at 24719 MOAP1 1124561 212530 at 24119 NEK7 1124563 212533 at 249441 WEE1 192 1124577 212552 at 3618 HPCAL1 1124583 212558 _at 20977 GDAPILl 1124594 1212572 at 184523 STK38L 1124606 212588 at 444324 PTPRC 1124610 212592 at 381568 IGJ 1124613 212599 _at 296720 AUTS2 1124616 212603 at 154655 MRPS31 1136759 212605 s at 188882 1124620 212610 at 83572 PTPN11 1136762 212624 s at 380138 CHN1 1136765 212629 s at 69171 PRKCL2 1124646 212646 at 436432 RAFTLIN 1136774 212657 s at 81134 ILIRN 1124655 212658 at 79299 LHFPL2 1124658 212663 at 522351 KIAA0674 1136777 212671 s at 387679 HLA-DQA1 1124666 212672 at 526394 ATM 1136781 212680 x at 120197 PPPIR14B 1136784 212689 s at 321707 JMJD1 1136786 212694 s at 63788 PCCB 1136788 212698 s at 355455 09/10/2004 1124692 212713 at 296049 MFAP4 1124705 212730 at 381347 DMN 1124712 212738 at 80305 ARHGAP19 1124713 212740 at 306747 PIK3R4 1124723 212753 at 435065 RNF3 1124733 212771 _at 66762 LOC221061 1124734 212774 at 446677 ZNF238 1124745 212789 at 438550 K1AA0056 1136819 212798_sat 112605 DKFZP5640 043 1124753 212801_at 528307 CIT 1124755 212805 at 23311 KIAA0367 1124760 212813 at 419149 JAM3 1124768 212824 at 98751 FUBP3 1124770 212827 at 153261 IGHM 1136831 212841 s at 12953 PPFIBP2 1136832 212842 x at 434959 RANBP2L1 1124782 212843 at 78792 NCAM1 1124786 212847 at 22370 NEXN 1124798 212867 at 446678 NCOA2 1124800 212871 at 413901 MAPKAPK5 1136844 212875 s at 16007 C21orf25 1124806 212881 at 105779 PIASY 1124820 212899 at 129836 CDK11 1124830 212911 at 9059 KIAA0962 1124831 212912 at 301664 RPS6KA2 1124833 212914 at 356416 CBX7 1136853 212922 s at 66170 SMYD2 1136859 212942 s at 212584 KIAA1199 1124862 212954 at 439530 DYRK4 1136865 212959 s at 412128 MGC4170 1124864 212960 at 411317 KIAA0882 1124875 212975 at 18166 KIAA0870 193 1124889 212993 at 244847 BTBD14A 1136876 212997 s at 445078 TLK2 1136877 212998 x at 409934 HLA-DQB1 1124893 213002 at 318603 MARCKS 1124913 213027 at 288178 SSA2 1124920 213039 at 6150 ARHGEF18 1124921 213044 at 306307 ROCK1 1124922 213045 at 173864 MAST3 1124941 213068 at 80552 DPT 1124942 213069 at 433452 HEG 1124948 213075 at 357004 LOC169611 1124953 213083 at 386278 SLC35D2 1136902 213086 s at 442592 1136903 213087 s at 334798 EEF1D 1124967 213108 at 143535 CAMK2A 1136913 213113 s at 99962 SLC43A3 1124972 213116 at 2236 NEK3 1136925 213154 s at 436939 BICD2 1125001 213158 at 16193 1125009 213169 at 27621 1125010 213170 at 43728 GPX7 1125013 213174 at 79170 TTC9 1136938 213188 s at 23294 MINA53 1136939 213193 x at 419777 1125025 213196 at 301094 1125027 213198 at 371974 ACVRIB 1125058 213238 at 437241 ATP1OD 1125079 213264 at 211601 MAP3K12 1125122 213324 at 436015 SRC 1125124 213326 at 20021 VAMPI 1136971 213330 s at 257827 STIP1 1136972 213331 s at 414410 NEKI 1125130 213338 at 35861 RISI 1125132 213341 at 47367 FEMIC 1125136 213348 at 106070 CDKNIC 1136983 213360 s at 450237 LOC340318 1136984 213364 s at 498154 SNX1 1136987 213370 s at 21695 SFMBT1 1136988 213373 s at 243491 CASP8 1136996 213397 x at 283749 RNASE4 1125181 213418 at 3268 HSPA6 1125195 213438 at 7309 1137022 213475 s at 174103 ITGAL 1125231 213489 at 446375 MAPRE2 1137026 213490 s at 366546 MAP2K2 1125245 213517 at 132977 PCBP2 1125246 213518 at 496511 PRKCI 1125249 213523 at 244723 CCNE1 1137042 213524 s at 432132 GOS2 1125279 213575 at 445652 TRA2A 1125305 213627 at 376719 MAGED2 1137097 213656 s at 20107 KNS2 1137109 213689 x at 469653 RPL5 1137112 213693 s at 89603 MUC1 194 1137137 213746 s at 195464 FLNA 1125377 213748 at 196966 KIAA0298 1125397 213784 at 415172 RABL4 1137158 213794 sat 9043 C14orfl2O 1137201 213877 x at 433343 SRRM2 1137202 213881 x at 380973 SMT3H2 1125456 213906 at 300592 MYBL1 1125459 213909 at 288467 LRRC15 1125462 213915 at 10306 NKG7 1125485 213958 at 436949 CD6 1137247 213975 s at 234734 LYZ 1137273 214020 x at 149846 ITGB5 1125516 214032 at 234569 ZAP70 1125520 214038 at 271387 CCL8 1137289 214049 x at 36972 CD7 1125527 214051 at 422848 MGC39900 1137291 214055 x at 446197 XTP2 1125532 214058 at 437922 MYCLI 1125546 214081 at 125036 PLXDC1 1137308 214093 s at 118962 FUBP1 1137328 214130 s at 502577 PDE4DIP 1137332 214146 s at 2164 PPBP 1137343 214170 x at 391168 FH 1125593 214180 at 8910 MANiCI 1137360 214196 s at 429658 CLN2 1137378 214228 x at 129780 TNFRSF4 1125634 214265 at 171025 ITGA8 1125658 214322 at 12436 CAMK2G 1137439 214339 s at 95424 MAP4KI 1137447 214359 s at 74335 HSPCB 1137449 214363 s at 223745 MATR3 1125685 214371 at 103978 STK22B 1137481 214428 x at 150833 C4A 1137486 214442 s at 441069 MiZi 1137488 214448 x at 9731 NFKBIB 1137492 214459 x at 274485 HLA-C 1125742 214470 at 169824 KLRB1 1137506 214501 s at 75258 H2AFY 1137512 214512 s at 229641 PC4 1137534 214551 s at 36972 CD7 1125789 214560 at 511953 FPRL2 1137539 214567 s at 458346 XCL2 1125818 214607 at 152663 PAK3 1125826 214617 at 2200 PRFI 1137561 214639 s at 67397 HOXA1 1125852 214660 at 439320 ITGA1 1125854 214663_at 6874 DustyPK 1137582 214683 s at 433732 CLKI 1137583 214687 x at 273415 ALDOA 1125872 214696 at 417157 MGC14376 1137594 214710 s at 23960 CCNBI 1137597 214721 x at 3903 CDC42EP4 1137601 214730 s at 78979 GLG1 1125901 214745 at 193143 KIAA1069 195 1125916 214764 at 497770 1125917 214769 at 417091 CLCN4 1125919 214772 at 432369 G2 1125921 214777 at 512003 1125927 214787 at 511742 IRLB 1125928 214790 at 435628 SUSPI 1137626 214797 s at 445402 PCTK3 1137643 214864 s at 155742 GRHPR 1137663 214909 s at 247362 DDAH2 1126047 214969 at 437214 MAP3K9 1137687 214974 x at 89714 CXCL5 1137698 215001 s at 442669 GLUL 1126081 215030 at 309763 GRSF1 1137742 215111 s at 114360 TSC22 1126131 215117 at 159376 RAG2 1137751 215127 s at 241567 RBMS1 1126148 215143 at 408264 FLJ36166 1137760 215158 s at 169681 DEDD 1137771 215193 x at 308026 HLA-DRB3 1137782 215223 s at 384944 S002 1137806 215313 x at 181244 HLA-A 1137809 215332 s at 405667 CD8B1 1126293 215346 at 504816 TNFRSF5 1137838 215411 s at 437508 C6orf4 1137868 215493 x at 169963 BTN2A1 1126387 215499 at 180533 MAP2K3 1126408 215528 at 22689 1137908 215603 x at 454906 1137955 215722 s at 434901 SNRPA1 1126540 215750 at 474916 KIAA1659 1126554 215767 at 159528 LCC91752 1126559 215776 at 248138 INSRR 1138030 215925 s at 116481 CD72 1138048 215967 s at 403857 LY9 1138120 216178 x at 287797 ITGB1 1138128 216199 s at 390428 MAP3K4 1138132 216207 x at 390427 IGKV1D-13 1138136 216215 s at 433574 RBM9 1138147 216234 s at 194350 PRKACA 1138150 216237 s at 77171 MCM5 1138157 216251 s at 82563 KIA0153 1126858 216261 at 87149 ITGB3 1138192 216321 s at 126608 NR3C1 1126892 216331 at 74369 ITGA7 1138244 216442 x at 418138 FN1 1138259 216484 x at 89525 HDGF 1138279 216520 s at 374596 TPT1 1138312 216598 s at 303649 CCL2 1138331 216640 s at 212102 P5 1138355 216705 s at 407135 ADA 1138379 216836 s at 446352 ERBB2 1127214 216837 at 201920 EPHA5 1138392 216862 s at 3548 MTCP1 1138400 216876 s at 41724 IL17 196 1138417 216905 s at 56937 ST14 1138421 216913 s at 434251 KIAA0690 1138441 216945 x_at 397891 PASK 1138443 216950 s at 77424 FCGRIA 1127290 217019 at 447032 1127294 217028 at 421986 CXCR4 1138507 217066 s at 898 DMPK 1138515 217080 s at 93564 HOMER2 1138532 217128 s at 199068 CAMK1G 1138537 217140 s at 1138538 217143 s at 2014 TRD@ 1138541 217149 x at 203420 TNK1 1127371 217164 at 391858 TIAI 1138555 217184 s at 434481 LTK 1138567 217200 x at 355264 CYB561 1138645 217373 x at 212217 MDM2 1138647 217377 x at 171262 ETV6 1138652 217388 s at 444471 KYNU 1138670 217422 s at 262150 C022 1138671 217427 s at 415735 HIRA 1138677 217436 x at 390440 1127567 217529_at 440667 FLJ20013 1127576 217544 at 529751 1138721 217552_x at 334019 CR1 1138759 217707 x at 396404 SMARCA2 1138765 217716 s at 306079 SEC61A1 1138778 217736 s at 434986 HRI 1138780 217739 s at 293464 PBEFI 1138783 217742 s at 370152 WAC 1138789 217750 s at 369120 FLJ13855 1127720 217765 at 272736 NRBP 1138801 217774 s at 333579 HSPC152 1127742 217814_at 8207 GKOO1 1127744 217817 at 323342 ARPC4 1138832 217829 s at 12820 USP39 1138845 217849 s at 436985 CDC42BPB 1127756 217850 at 313544 NS 1127761 217863 at 75251 PIAS1 1138858 217871 s at 407995 MIF 1127775 217886 at 79095 EPS15 1138867 217892 s at 10706 EPLIN 1138874 217910_xat 383019 TCFL4 1138878 217917 s at 100002 DNCL2A 1138887 217937sat 200063 HDAC7A 1127805 217947_at 380627 CKLFSF6 1127807 217950, at 7236 NOSIP 1127813 217962_at 14317 NOLA3 1138905 217970_sat 437844 KIAA1 194 1127822 217977 at 279623 SEPX1 1138910 * 217982_sat 374503 MORF4L1 1127833 218001_ at 382044 MRPS2 1138920 218002 s at 24395 CXCL14 1127838 218012 at 136164 SE20-4 1127849 218032 at 76691 SNN 197 1138944 218051 s at 84753 FLJ12442 1127864 218066 at 172613 SLC12A7 1138959 218076 s at 203605 RICH1 1127873 218089 at 11314 C20orf4 1138973 218097 s at 11270 ClOorf66 1127885 218113 at 160417 TMEM2 1138994 218143 s at 238030 SCAMP2 1138995 218144 s at 24956 FLJ22056 1127901 218145 at 344378 C20orf97 1139005 218168 s at 273186 CABC1 1139017 218189 s at 274424 NANS 1139026 218205 s at 512094 MKNK2 1127931 218208 at 288284 PQLC1 1139037 218223 s at 173380 CKIP-1 1127940 218227 at 256549 NUBP2 1139039 218228 s at 280776 TNKS2 1127943 218232 at 9641 CIQA 1139048 218250 s at 170553 CNOT7 1139054 218263 s at 25726 LOC58486 1139076 218306 s at 133411 HERC1 1139100 218350 s at 234896 GMNN 1139105 218367 x at 8015 USP21 1139106 218368 s at 355899 TNFRSF12A 1139127 218409 s at 13015 DNAJC1 1128042 218436 at 297875 SILl 1128066 218475 at 63609 HTF9C 1128070 218481 at 283741 RRP46 1128079 218499 at 23643 MST4 1128095 218520 at 432466 TBK1 1128099 218529 at 333427 8D6A 1128100 218530_at 95231 FHOD1 1139185 218535 s at 27021 RIOK2 1128106 218542 at 14559 Cl0orf3 1128111 218552 at 170915 FLJ10948 1139196 218559 s at 169487 MAFB 1139202 218569_s at 440695 KBTBD4 1128125 218581 at 445665 ABHD4 1139215 218597 s at 43549 ClOorf7O 1128144 218613_at 236438 DKFZp761K _________ 1423 1128151 218625 at 103291 NRNI 1128157 218631 at 23918 VIP32 1139226 218633 x at 266514 FLJ11342 1139230 218640 s at 29724 PLEKHF2 1128164 218646 at 44344 FLJ20534 1139235 218651 s at 416755 FLJI1196 1128167 218653 at 78457 SLC25A15 1128174 218665 at 19545 FZD4 1128192 218696 at 102506 EIF2AK3 1128195 218699 at 115325 RAB7L1 1139265 218722 s at 187657 FLJ12436 1139266 218723 s at 76640 RGC32 1128214 218734 at 408443 FLJ13848 1139274 218740 s at 20157 CDK5RAP3 198 1139277 218747 s at 267993 TAPBP-R 1139280 218751 s at 312503 FBXW7 1128223 218753 at 55024 FLJ10307 1128231 218764 at 315366 PRKCH 1139301 218792 s at 108502 BSPRY 1139303. 218794 s at 134406 FLJ20511 1128248 218802 at 234149 FLJ20647 1139314 218831 s at 111903 FCGRT 1128283 218856 at 159651 TNFRSF21 1128287 218862 at 300063 ASB13 1128298 218887 at 55041 MRPL2 1128311 218909 at 30352 RPS6KC1 1128321 218921 at 433036 SIGIRR 1139360 218947 s at 173946 FLJ10486 1128341 218955 at 274136 BRF2 1128356 218983 at 415792 C1RL 1128360 218988 at 445043 SLC35E3 1128377 219014 at 371003 PLAC8 1128386 219025 at 195727 CD164L1 1128387 219028 at 397465 HIPK2 1139393 219032 x at 170129 'OPN3 1128395 219039 at 7188 SEMA4C 1128401 219049 at 341073 ChGn 1139411 219073 s at 368238 OSBPL10 1128418 219082 at 433499 CGI-14 1128435 219109 at 6783 PF20 1128439 219118 at 438695 FKBP11 1128447 219130 at 40337 FLJ10287 1128457 219148 at 104741 TOPK 1139444 219151 s at 355874 RABL2B 1128469 219173 _at 390817 FLJ22686 1128471 219176 at 3592 FLJ22555 1139461 219191 s at 14770 BIN2 1128494 219209 at 389539 MDA5 1139466 219210 s at 365655 RAB8B 1128506 219226 at 416108 CRK7 1139483 219249 s at 3849 FKBP10 1128535 219278 at 194694 MAP3K6 1128536 219279 at 21126 DOCK10 1139526 219356 s at 415534 C9orf83 1139528 219360 s at 31608 TRPM4 1139531 219365 s at 145156 MGC8407 1128585 219366 at 63168 AVEN 1139542 219396 s at 512732 NEILl 1128615 219410 at 104800 FLJ10134 1128626 219424 at 501452 EB13 1139552 219441 s at 413386 LRRK1. 1128648 219452 at 499331 DPEP2 1139556 219457 s at 413374 RIN3 1128653 219461 at 21420 PAK6 1128655 219463 at 22920 C200rf103 1128660 219471 at 413071 C13orfl8 1128681 219500 at 191548 CLC 1128688 219509 at 238756 MYOZ1 199 1139572 219511 s at 24948 SNCAIP 1128694 219517_at 171466 ELL3 ,1139575 219519 s at 31869 SN 1139579 219528 s at 57987 BCL11B 1128705 219535 at 109437 HUNK 1128710 219542 at 159146 NEKI1 1128713 219545 at 17296 KCTD14 1128733 219572 at 489847 CADPS2 1128738 219581 at 335550 MGC2776 1139603 219603 s_at 145956 ZNF226 1128757 219618 at 142295 IRAK4 1128781 219648 at 79741 FLJ10116 1128786 219654 at 114062 PTPLA 1128787 219655 at 114611 C7orflO 1139623 219667 s at 193736 BANK1 1128801 219676 at 288539 ZNF435 1128807 219686 at 58241 HSA250839 1128845 219734 at 272416 FLJ20174 1128860 219753 at 323634 STAG3 1139645 219757 s at 134051 C14orflOl 1139654 219787 s at 293257 ECT2 1139661 219806 s at 416456 FN5 1128900 219812 at 323634 STAG3 1128901 219813 at 487239 LATS1 1139663 219816 s at 4997 RNPC4 1128915 219831 at 105818 CDKL3 1139669 219837 s at 13872 C17 1128965 219901 at 170623 FGD6 1128969 219906 at 446590 FLJ10213 1129024 220005_at 13040 GPR86 11'29026 220007_at 135146 FLJ13984 1129043 220028 _at 23994 ACVR2B 1129049 220034 at 268552 IRAK3 1129059 220054 at 98309 IL23A 1129061 220056 at 110915 IL22RA1 1129064 220059 at 121128 BRDG1 1129071 220068_at 136713 VPREB3 1129085 220088 at 2161 C5RI 1129103 220118 at 99430 TZFP 1139767 220127 s at 12439 FBXL12 1139774 220140 s at 15827 SNX11 1129120 220146 at 179152 TLR7 1129151 220196 at 432676 MUC16 1139805 220230 s at 414362 CYB5R2 1129203 220273 at 110040 IL17B 1129223 220296 at 13785 GALNTIO 1129228 220302 at 148496 MAK 1129232 220307 at 157872 CD244 1129245 220322 at 211238 ILIF9 1139830 220330 s at 221851 SAMSN1 1139831 220335 x at 268700 FLJ21736 1129265 220351 at 310512 CCRL1 1139839 220357 s at 62863 SGK2 1129269 220358 at 62919 SNFT 200 1139842 220367 s at 133523 SAP13d 1129281 220377 at 395486 C14orfl10 1129310 220415 at 414091 TNNI3K 1129336 220448 at 252617 KCNK12 1129419 220565 at 278446 GPR2 1139925 220643 s at 173438 FAIM 1129495 220684 at 272409 TBX21 1129517 220712 at 1139949 220725 x at 528684 FLJ2355B 1139950 220731 s at 437385 FLJ10420 1129535 220737 at 368153 RPS6KAS 1139955 220740 s at 4876 SLC12A6 1139957 220742 s at 63657 NGLY1 1129537 220745 at 71979 1L19 1139962 220751 s at 10235 C5orf4 1139969 220761 s at 12040 JIK 1139971 220765 s at 127273 LIMS2 1140007 220865 s at 279865 TPRT 1140018 220917 s at 438482 PWDMP 1140027 220933 s at 12742 ZCCHC6 1140031 220937 s at 3972 SIAT7D 1129661 220971 at 302036 IL17E 1140072 220984 s at 199750 SLCO5A1 1140075 220987 s at 172012 SNARK 1140088 221002 s at 509050 DC-TM4F2 1140127 221044 s at 125300 TRIM34 1140151 221080 s at 236449 FAM31C 1129681 221085 at 241382 TNFSF15 1129694 221111 at 272350 IL26 1129743 221191_at 429531 DKFZP434A 0131 1140214 221215 s at 55565 ANKRD3 1140236 221239 s at 194976 SPAP1 1140238 221241 s at 11962 BCL2LI4 1129754 221271 at 302014 IL21 1129760 221287 at 404277 RNASEL 1129812 221355 at 248101 CHRNG 1129821 221367_at 248146 MOS 1129825 221371 at 248197 TNFSF18 1129874 221463_at 247838 CCL24 1129879 221468 at 248116 XCRI 1140344 221479 s at 132955 BNIP3L 1129887 221485 at 107526 B4GALT5 1140370 221520 s at 48855 CDCA8 1140378 221530 s at 437282 BHLHB3 1129907 221539 at 406408 EIF4EBP1 1129911 221549 at 400625 GRWO1 1140391 221558 s at 44865 LEF1 1129917 221560 at 118843 MARK4 1129923 221571 at 297660 TRAF3 1140399 221577 x at 296638 GDF1 5 1140404 221584 s at 354740 KCNMA1 1140416 221601 s at 58831 TOSO 1129943 221626 at 512828 ZNF506 201 1140457 221658 s at 210546 IL21 R 1140464 221667 s at 111676 HSPB8 1140473 221676 s at 17377 CORO1C 1140491 221696_sat 24979 DKFZp761P 1010 1140497 221704 s at 77870 FLJ12750 1129967 221739 at 10927 Cl9orf10 1140520 221741 s at 11747 C20orf2l 1129978 221753 at 60377 SSH1 1140524 221766 s at 10784 C6orf37 1129993 221777 at 412981 FLJ14827 1140534 221790 s at 184482 ARH 1130007 221796 at 439109 NTRK2 1130030 221834 at 301872 LONP 1130040 221855 at 356460 1130054 221872 at 82547 RARRESI 1140565 221875 x at 411958 HLA-F 1140567 221881 s at 25035 CLIC4 1140570 221891 x at 180414 HSPA8 1140571 221893 s at 210397 ADCK2 1130072 221898 at 468675 T1A-2 1130078 221905 at 386952 CYLD 1140574 221912 s at 17987 MGC1203 1130088 221918 at 258536 PCTK2 1130090 221922 at 278338 GPSM2 1140584 221932 s at 294083 C14orf87 1140589 221942 s at 433488 GUCY1A3 1130114 221965 at 445084 MPHOSPH9 1130117 221969 at 22030 PAX5 1130121 221978 at 411958 HLA-F 1.140613 221998 s at 443330 VRK3 1140630 222033 s at 347713 FLTI 1140632 222036 s at 460184 MCM4 1130155 222043 at 436657 CLU 1130168 222061 at 75626 CD58 1130169 222062 at 132781 IL27RA 1130201 222126 at 278502 HRBL 1140729 222223 s at 207224 IL1F5 1140745 222245 s at 72222 FER1L4 1130293 222315 at 292853 1130337 222368 at 491069 1095985 222450 at 83883 TMEPA 1095996 222482 at 288801 SSBP3 1114679 222503 s at 16470 FLJ10904 1096028 222557_ at 285753 STMN3 1114715 222565 s at 434387 PRKCN 1096035 222569 at 105794 UGCGLI 1096038 222572 at 22265 PPM2C 1114726 222590 s at 3532 NLK 1096054 222606 at 21331 FLJ10036 1096070 222640 at 241565 DNMT3A 1096077 222659 at 441043 IPO11 1096078 222661 at 284216 HSU84971 1114766 222666 s at 113052 RCLI 202 1096085 222674 at 224137 HSPC1O9 1096108 222731 at 292871 ZDHHC2 1114824 222762 x at 193370 LIMD1 1114853 222812 s at 512618 ARHF 1096149 222824 at 410205 NUDT5 1096152 222828 at 288240 IL20RA 1096158 222838 at 132906 SLAMF7 1096163 222848 at 164018 FKSG14 1114877 222862 s at 18268 AK5 1096172 222880 at 300642 AKT3 1096180 222890 at 11614 HSPC065 1114893 222891 s at 314623 BCL11A 1096182 222899 at 256297 ITGA11 1114913 222920 s at 33187 KIAA0748 1096220 222974 at 287369 IL22 1114967 223028 s at 7905 SNX9 1114970 223032 x at 279529 ,PX19 1096248 223040 at 109253 NAT5 1096251 223044 at 409875 SLC40A1 1114977 223052 x at 30026 HSPCI82 1114981 223057 s at 203206 XPO5 1114988 223075 s at 4944 C9orf58 1115008 223117 s at 441028 USP47 1115012 223122 s at 31386 SFRP2 1096297 223141 at 9597 UCKI 1096300 223151 at 74284 MGC2714 1115034 223158 s at 387222 NEK6 1115052 223190 s at 380021 MLL5 1115071 223218 s at 390476 MAIL 1115073 223220 s at 131315 BAL 1096341 223241 at 12169 SNX8 1096356 223266 at 259230 ALS2CR2 1096357 223267 at 57898 FLJ20432 1096362 223274 at 512706 TCF19 1096364 223276 at 29444 NID67 1096369 223286 at 417029 DERP6 1096378 223303 at 180535 URP2 1096379 223304 at 439590 SLC37A3 1115128 223349 s at 293753 BOK 1096406 223361 at 238205 C6orf115 1096429 223405 at 64896 NPL 1115160 223413 s at 425427 LYAR 1096440 223423 at 231320 GPR160 1096442 223430 at 306864 SIK2 1096446 223434 at 92287 GBP3 1115176 223451 s at 15159 CKLF 1096456 223454 at 82407 CXCL16 1096460 223460 at 8417 CAMKK1 1096466 223467 at 25829 RASD1 1096469 223471 at 103267 RAB31P 1115186 223480 s at 283734 MRPL47 1115194 223502 s at 270737 TNFSFI3B 1096499 223514 at 293867 CARDl 1 1096503 223522 at 21379 C9orf45 203 1115203 223534 s at 414481 RPS6KL1 1096530 223565 at 409563 PACAP 1115226 223600 s at 279718 KIAA1683 1096570 223624 at 409813 ANUBL1 1096579 223640 at 117339 HCST 1115253 223664 x at 310922 BCL2L13 1096609 223696 at 528631 ARSD 1115271 223705_s_at 71252 DKFZp761C 169 1096615 223707 at 356342 RPL27A 1096616 223708 at 119302 C1QTNF4 1096617 223710 at 131342 CCL26 1096621 223715 at 170819 STK29 1115286 223750 s at 120551 TLR10 1115290 223759 s at 193666 GSG2 1115303 223787 s at 236257 LOC51244 1115309 223804_s_at 443081 DKFZP434F 091 1096690 223827 at 334174 TNFRSF19 1096693 223834 at 443271 PDCD1LG1 1115329 223852 s at 439658 MGC4796 1096719 223874 at 250153 ARP3BETA 1115338 223883 s _at 224355 STK31 1096738 223903 at 87968 TLR9 1115347 223909 s at 112272 HDAC8 1096742 223910 at 114905 ERN2 1115360 223940_ x at 187199 PRO1073 1096805 224027 at 334633 CCL28 1096829 224071 at 272373 IL20 1096834 224079 at 278911 IL17C 1096877 224132 at 326732 MGC13008 1115441 224156 x at 5470 IL17RB 1096903 224185 at 437460 FLJ10385 1096936 224262 at 306974 IL1FlO 1115519 224302 s at 408914 MRPS36 1096965 224346 at 433466 PRO1853 1115566 224369 s at 163825 SP329 1096981 224399 at 61929 PDCD1LG2 1115587 224402 s at 120260 IRTAI 1115589 224406 s at 415950 IRTA2 1115591 224409 s at 367871 SSTK 1115607 224428 s at 435733 CDCA7 1115621 224450 s at 437474 RIOK1 1115646 224481 s at 210850 HECTD1 1115668 224509 s at 155839 RTN41P1 1115673 224514 x at 129959 ILl 7RC 1115679 224523 s at 8345 MGC4308 1115695 224553 s at 212680 TNFRSF18 1115696 224555 x at 166371 ILIF7 1115704 224569 s at 350268 IRF2BP2 1097030 224574 at 511801 1097065 224621 at 324473 MAPK1 1097096 224659 at 8518 SEPN1 1097107 224673 at ,502378 LENG8 204 1097109 224675 at 78871 MESDC2 1097126 224694 at 274520 ANTXR1 1097143 224716 at 74335 HSPCB 1097156 224733 at 298198 CKLFSF3 1097161 224740 at 5064 1097172 224753 at 434886 CDCA5 1097177 224761 at 9691 GNA13 1097195 224785 at 149931 MGC29814 1097202 224796 at 386779 DDEF1 1097229 224830 at 446393 CPSF5 1097236 224837 at 235860 FOXP1 1115763 224839 s at 355862 GPT2 1097247 224851 at 388761 CDK6 1097253 224859 at 77873 B7H3 1097255 224861 at 380144 1097271 224880 at 6906 RALA 1097280 224891 at 423523 1097281 224892 at 7037 PLDN 1097282 224893 at 356719 LOC283241 1097290 224903 at 151001 CIRH1A 1097297 224917 at 166254 VMP1 1097307 224929 at 379754 LOC340061 1097310 224934 at 5672 SMAP-5 1097325 224951 at 458450 LASS5 1097329 224955 at 528675 TEADI 1097334 224960at 71573 FLJ10074 1097359 224990_at 518723 1097365 224998 at 325825 CKLFSF4 1097371 225005 at 7299 PHF13 1097383 225019 at 111460 CAMK2D 1097388 225024 at 278839 C20orf77 1097395 225032 at 299883 FAD104 1115800 225040 s at 282260 RPE 1097424 225067_at 7978 DKFZP434C 131 1097441 225086 at 6799 FLJ38426 1097448 225093_at 250607 UTRN 1115812 225164 s at 412102 EIF2AK4 1115813 225175 s at 105509 CTL2 1097540 225195 at 388087 1097553 225214 at 197071 PSMB7 1097561 225224_at 19221 DKFZP566G 1424 1097563 225226 at 169577 FLJ14743 1097564 225227 at 272108 SKIL 1115829 225253 s at 433213 METTL2 1097600 225272 at 10846 SAT2 1097609 225283 at 6093 ARRDC4 1097610 225284 at 6019 DNAJC3 1097611 225285 at 438993 BCAT1 1097614 225289 at 410491 MGC16063 1115840 225308 s at 437362 KIAA1728 1097637 225317 at 63220 ACBD6 1097665 225351 at 434241 HT011 205 1097676 225366 at 23363 PGM2 1097683 225373 at 132569 PP2135 1097684 225374 at 368878 MGC45714 1097704 225399 at 440663 Clorf19 1097707 225402 at 440263 C20orf64 1097717 225412 at 23317 FLJ14681 1097735 225436 at 26765 LOC58489 1097804 225519_at 446590 FLJ10213 1097814 225529 at 21446 CENTB5 1115876 225535 s at 11866 TIMM23 1097824 225540 at 167 MAP2 1115877 225552 x at 76239 MGC3047 1097887 225611 at 212787 KIAA0303 1097897 225622_at 266175 PAG 1097899 225624 at 145047 LOC92017 1097901 225626_at 266175 PAG 1115888 225629 s at 35096 ZBTB4 1097902 225630 at 412318 KIAA1706 1115892 225649 s at 100057 STK35 1097918 225650 at 140309 LOC90378 1097928 225660_ at 443012 SEMA6A 1097930 225662 at 115175 ZAK 1097940 225673 at 380906 MYADM 1115895 225682 s at 202505 RPC8 1097948 225684 at 69476 LOC348235 1097961 225699 at 25892 1097966 225704 at 127270 KIAA1 545 1097976 225715 at 218017 raptor 1098012 225756 at 355669 CSNKIE 1115905 ' 225757 s at 301478 CLMN 1098023 225773 at 181161 KIAA1 972 1098065 225817 at 10119 FLJ14957 1098069 225823 at 356626 1115916 225836 s at 157148 MGC13204 1115917 225849 s at 284265 C6orf83 1098095 225852_at 131059 ANKRD17 1098103 225864 at 124951 NSE2 1098145 225913 at 9587 KIAA2002 1098152 225922 at 377588 KIAA1450 1098156 225927 at 170610 MAP3K1 1098168 225943 at 22151 NLN 1098174 225949 at 274401 LOC340371 1098179 225956 at 163725 LOC153222 1098186 225964 at 288697 MGC11349 1098195 225974_at 88594 DKFZp762C 1112 1098204 225984 at 43322 PRKAA1 1098220 226002 at 80720 GAB1 1098234 226016 at 446414 CD47 1098235 226017 at 440494 CKLFSF7 1098242 226025 at 273104 KIAA0379 1098252 226035 at 16953 USP31 1098256 226041 at 431871 SVH 1098258 226043 at 239370 GPSM1 206 1098268 226053 at 110299 MAP2K7 1098271 226056 at 300670 CDGAP 1098277 226065 at 6786 PRICKLE1 1098278 226066_at 166017 MITF 1098303 226096 at 15463 FNDC5 1115953 226111 s at 278422 ZNF385 1115955 226132 s at 7988 FLJ31434 1098338 226136 at 269857 HRB2 1115960 226145 s at 15420 FRAS1 1115965 226166 x at 26996 STK36 1098405 226218 at 362807 IL7R 1098412 226225 at 409515 MCC 1098415 226230 at 130900 KIAA 387 1098433 226250 at 202577 1098447 226267 at 154095 JDP2 1098459 226279 at 25338 SPUVE 1098461 226281 at 234074 DNER 1098476 226299 at 300485 pknbeta 1098495 226318 at 443668 TBRG1 1098506 226333 at 193400 IL6R 1098521 226350 at 170129 OPN3 1098548 226377 at 436639 NFIC 1098550 226381 at 355655 1098553 226384 at 437179 HTPAP 1098574 226410 at 79077 KIAA0233 1098592 226431_at 283707 ALS2CR13 1098604 226444 at 32793 SLC39A10 1098607 226448_at 38516 MGC15887 1098611 226452_at 433611 PDKI 1098613 226454 at 388125 RENT1 1098618 226459_at 374836 PIK3AP1 1116001 226465 s at 430541 SON 1098629 226473 at 103305 1116006 226491 x at 172550 PTBP1 1098658 226507 at 64056 PAK1 1098668 226517 at 438993 BCAT1 1098669 226518 at 302746 KCTD1O 1098678 226530 at 386140 BMF 1098683 226535 at 57664 ITGB6 1098694 226548 at 97837 1098718 226574 at 16364 PSPC1 1116022 226611 s at 433422 p30 1098771 226638_at 374446 KIAA 501 1098784 226653 at 12808 MARK1 1098809 226682 at 359394 1098821 226694 at 42322 PALM2 1098822 226695 at 443452 PRRX1 1098832 226705 at 748 FGFR1 1098840 226713 at 55098 C3orf6 1098862 226737 at 303669 MGC26694 1098865 226741_at 250905 LOC51234 1098883 226760 at 412014 MBTPS2 1098893 226771 at 43577 ATP8B2 1098898 226777 at 8850 ADAM 12 207 1098909 226789 at 446408 1098918 226799 at 170623 FGD6 1098927 226811 at 356216 FLJ20202 1116045 226828 s at 23823 HEYL 1098946 226834 at 135121 ASAM 1098951 226840 at 75258 H2AFY 1098952 226841 at 62264 KIAA0937 1098954 226844 at 128905 MOBKL2B 1098962 226853 at 20137 BMP2K 1098978 226869 at 124863 1098987 226879 at 412559 FLJ21127 1098991 226884 at 126085 LRRN1 1116056 226913 s at 243678 SOX8 1099028 226930 at 334838 FNDC1 1099032 226936 at 35962 1099040 226944 at 390421 HTRA3 1116063 226957 x at 75447 RALBP1 1099053 226959 at 376041 1099058 226964 at 425116 1099072 226979 at 28827 MAP3K2 1099088 226996 at 14355 1099105 227013 at 78960 LATS2 1099112 227020 at 368672 1099120 227030 at 371680 1099124 227034 at 355455 09/10/2004 1099128 227039 at 350631 AKAP13 1099135 227046 at 3402 SLC39A11 1099140 227052 at 500350 1099148 227060 at 434975 TNFRSF19L 1099150 227062 at 240443 1099152 227064 at 351247 MGC15396 1099154 227066 at 97927 MOBKL2C 1116071 227067 x at 502564 FLJ20719 1099167 227080 at 381105 MGC45731 1116073 227103 s at 146161 MGC2408 1099204 227121 at 193784 1116085 227173 s at 88414 BACH2 1099265 227193 at 375762 1099291 227222 at 130774 FBXOIO 1099292 227223 at 282901 RNPC2 1099299 227232 at 241471 EVL 1099318 227255 at 29911 LOC149420 1099328 227267 at 432726 FLJ35779 1099332 227272 at 32433 1099358 227300 at 93135 1116103 227308 x at 289019 LTBP3 1099377 227324 at 130712 ADCK4 1099388 227336 at 124024 DTX1 1099396 227346 at 435949 ZNFN1A1 1099403 227354 at 266175 PAG 1099418 227370 at 172792 KAA1946 1099444 227407 at 434489 FLJ90013 1116122 227408_s_at 42768 DKFZp7610 ____ ___ ___ ___ ____ ___ ___ ___ ___ ___ ___ ___ 0113 208 1116126 227432 s at 438669 INSR 1099510 227482 at 15251 ADCK1 1099526 227502 at 521240 LCHN 1099539 227520 at 201624 CXorf15 1099549 227533 at 446665 1099563 227550 at 388347 1099598 227590 at 511859 1116150 227606 s at 16229 AMSH-LP 1099631 227624 at 367639 FLJ20032 1099633 227627 at 380877 SGKL 1099651 227646 at 120785 EBF 1099669 227666 at 45057 MGC45428 1099680 227677 at 210387 JAK3 1099686 227684 at 117721 1099699 227697 at 436943 SOCS3 1099711 227713 at 243596 1099734 227740 at 127310 KIS 1099743 227750 at 162189 TRAD 1099748 227755 at 356481 1099760 227767 at 129206 CSNKIG3 1099798 227811 at 411081 FGD3 1099826 227842 at 445862 RAB30 1099830 227847 at 28020 EPM2AIP1 1099847 227867 at 36723 LOCI 29293 1099857 227877 at 119768 1116181 227891 s at 402752 TAF15 1099886 227917 at 511708 1099900 227934 at 444508 1099939 227983 at 488173 MGC7036 1099951 227999 at 157728 LOC170394 1099953 228001 at 433668 C21orf4 1099960 228008 at 144583 1099965 228014 at 71962 L3C138428 1099978 228035 at 148135 STK33 1116219 228056 s at 322854 NAPIL 1099995 228057 at 107515 DDIT4L 1100005 228069 at 121536 DUFD1 1100027 228094 at 16291 AMICA 1100040 228109 at 410953 RASGRF2 1100042 228113 at 351413 RAB37 1116233 228128 x at 440769 PAPPA 1100054 228130 at 125353 1100060 228139 at 268551 RIPK3 1100071 228153 at 432653 IBRDC2 1100130 228224 at 76494 PRELP 1100136 228231 at 413078 NUDTI 1100138 228234 at 278391 T(RP 1100144 228240 at 436379 1100150 228248 at 9343 MGC39830 1100159 228258 at 32156 RPS6KB2 1100161 228261 at 135805 L3C142678 1100171 228273 at 528654 FLJ11029 1100183 228286 at 180582 FLJ40869 1100249 228367 at 388674 HAK 209 1100258 228377 at 88442 KIAM384 1100263 228382 at 406335 LOC90268 1116277 228384 s at 118210 ClOorf33 1100288 228411 at 26981 ALS2CR19 1100290 228414 at 4241 1100301 228426 at 356250 LLT1 1100311 228437 at 445890 HSPC163 1100335 228464 at 268474 1100339 228468 at 276905 MASTL 1100384 228524 at 283374 ADCKS 1100405 228549 at 119387 KAA0792 1100420 228565 at 50883 KIAA1 804 1100423 228568 at 50841 FLJ30973 1100433 228580 at 390421 HTRA3 1100443 228592 at 438040 MS4A1 1100496 228654 at 111496 LOC139886 1116317 228661 s at 526415 1100538 228709 at 432458 PRG4 1100561 228736 at 194109 HEL308 1100562 228737 at 26608 C200rfl00 1100581 228758 at 155024 BCL6 1100585 228762 at 159142 LFNG 1100591 228769.at 388162 HKR2 1100598 228776 at 531058 1100609 228788 at 447045 PPIL2 1100625 228806 at 232803 1100721 228918 at 18713 1100750 228955 at 280387 1100753 228958 at 512717 ZNF19 1100770 228976 at 65578 1100847 229070 at 97411 C6orflO5 1100849 229072 at 184430 1100851 229074 at 55058 EHD4 1100871 229101 at 48353 1100873 229103 at 445884 1100879 229111 at 119983 MASP2 1100904 229145 at 426296 LOC119504 1100911 229152 at 320147 C4orf7 1100916 229158_at 105448 PRKWNK4 1100977 229233 at 444783 NRG3 1100995 229256 at 26612 PGM2L1 1101004 229265 at 2969 SKI 1101023 229288 at 73962 EPHA7 1101054 229322 at 173328 PPP2R5E 1116432 229356 x at 409362 -KAA1259 1101096 229373 at 527236 1101119 .229401 at 390823 IL17RE 1101128 229411 at 436667 MGC45419 1116445 229436 x at 301927 C6.1A 1101149 229437 at 517226 BIC 1101211 229513 at 287659 STRBP 1101272 229584_at 179089 DKFZp434H 2111 1101276 229588 at 1098 ERdj5 2 10 1101291 229606 at 272458 PPP3CA 1101295 229610 at 99807 FLJ40629 1101305 229623 at 112742 1101322 229645 at 227699 1101354 229686 at 111377 P2RY8 1101416 229764 at 338851 FLJ41238 1101430 229779 at 418040 1101439 229790 at 63335 TERF2 1101477 229838 at 423095 NUCB2 1101478 229839 at 146246 MGC45780 1101514 229886 at 88801 FLJ32363 1101566 229947 at 98558 1101582 229967 at 195685 CKLFSF2 1101586 229971 at 187884 GPR114 1101628 230021 at 441708 MGC45866 1101634 230028 at 510588 1101687 230086 at 440808 FNBP1 1101708 230110 at 459526 MCOLN2 1101758 230170 at 248156 OSM 1101775 230191 at 343820 TTBK1 1101777 230193 at 359981 MGC33630 1101829 230252 at 155538 GPR92 1101892 230327 at 225948 1116593 230329 s at 422889 NUDT6 1101905 230345 at 170843 1101944 230391 at 439064 1101948 '230395 at 14411 1101974 230425 at 272311 EPHB1 1102027 230489 at 58685 CD5 1102030 230494 at 110855 SLC20A1 1102081 230551 at 506977 1102165 230650 at 152460 1102193 230680 at 22668 1102282 230788 at 934 GCNT2 1116666 230803_s_at 442801 DKFZP564B 1162 1102350 230864 at 25845 MGC42105 1116676 230894 s at 185084 MSI2 1102408 230934 at 306327 RAB3GAP 1102415 230942 at 99272 CKLFSF5 1102437 230966 at 437023 IL411 1102470 231007 at 292915 1102471 231008 at 158357 UNC5CL 1102479 231017 at 301772 STK11 1102537 231087 at 202151 1102540 231093 at 434881 FCRH3 1116715 231149 s at 123427 FLJ20574 1102633 231198 at 511124 1102652 231219 at 343717 CKLFSF1 1102654 231221 at 380599 KIAA0350 1102725 231303 at 234016 C21orf42 1102744 231324 at 198671 1102821 231412 at 202024 1102859 231455 at 446195 2 1 1 1102885 231481 at 130310 CCNB3 1102898 231496 at 145519 FKSG87 1102912 231514_at 194610 MGC15882. 1103054 231690 at 341531 1103107 231759 at 247978 TAL2 1103111 231763 at 436896 RPC155 1103120 231775 at 401745 TNFRSF10A 1103124 231779 at 424542 IRAK2 1103134 231792 at 86092 MYLK2 1103137 231796 at 283613 EPHAB 1103139 231798 at 248201 NOG 1116826 231823 s at 26204 KIAA1295 1116829 231840 x at 115467 LOC90624 1103224 231906 at 301963 HOXD8 1116844 231920 s at 405789 CSNK1G1 1103264 231954_at 142307 DKFZP43410 S~714 1103272 231964 at 137206 1103284 231978 at 186655 TPCN2 1116854 231992 x at 438623 1103303 232000 at 49605 C9orf52 1103304 232001 at 46919 1116863 232068 s at 174312 TLR4 1103390 232103 at 271752 BPNT1 1103398 232112 at 220745 FLJ10244 1103420 232138 at 372571 MBNL2 1116879 232160 s at 325630 TNIP2 1103475 232204 at 120785 EBF 1103497 232231 at 50115 1103504 232239 at 142517 1103540 232282 at 92423 PRKWNK3 1103639 232399 at 388304 KIAA1765 1103711 232478 at 288718 1103766 232546 at 192132 TP73 1103855 232645 at 259625 LC153684 1103858 232648 at 246240 PSMA3 1116958 232693 s at 27410 PBF 1103921 232724 at 371612 MS4A6A 1103932 232741 at 31330 1116966 232744 x_at 301124 1103982 232798 at 142926 MGC26226 1104072 232906 at 287429 1104175 233029 at 287383 KIAA1639 1104195 233052 at 172101 DNAH8 1117023 233110 s at 289052 BCL2L12 1104254 233121 at 492700 1104373 233271 at 1104545 233476 at 254477 1104552 233483 at 193857 LOC96597 1104840 233867 at 482250 1104870 233916 at 210958 K1A1486 1117211 233955 x at 356509 HSPC195 1104905 233964 at 13453 FLJ14753 1104910 233969 at 458262 IGL@ 2 12 1105001 234088 at 527386 1117245 234107 s at 527974 HARS2 1105178 234284 at 283961 GNG8 1117278 234312 s at 14779 ACAS2 1117298 234366 x at 449586 1105248 234403 at 1117343 234643 x at 306812 BUCS1 1117350 234672 s at 435982 FLJ10407 1117373 234725 s at 416077 SEMA4B 1117394 234792 x at 1117403 234863 x at 272027 FBXO5 1105668 234954 at 1105684 234973 at 195155 SLC38A5 1105728 235022 at 13034 MGC24180 1105732 235026 at 396626 FLJ32549 1105751 235046 at 176376 1105759 235056 at 171262 ETV6 1105798 235099 at 154986 CKLFSF8 1105814 235117 at 105223 1105832 235136 at 306777 GSDML 1105838 235142 at 129837 ZBTB8 1105842 235146 at 173392 KIAM 145 1105854 235158 at 267245 FLJ14803 1105866 235170 at 9521 ZNF92 1105900 235211 at 525015 1105915 235229 at 332649 1105935 235251 at 444290 1105936 235252 at 276238 KSR 1105959 235278 at 399982 1105986 235310 at 49614 GCET2 1106013 235341 at 6019 DNAJC3 1106015 235343 at 96885 FLJ12505 1106025 235353 at 49500 KIAA0746 1106030 235359 at 162185 UNQ3030 1106043 235372 at 266331 FREB 1106053 235383 at 154578 MYO7B 1106088 235421 at 499235 1106110 235444 at 235860 FOXPI 1106124 235458 at 155111 HAVCR2 1106126 235460 at 434937 PPIB 1106159 235496 at 208081 1106196 235536 at 142074 1106204 235545 at 445098 SDP35 1106230 235572 at 381225 Spc24 1106279 235626 at 130065 CAMK1D 1106306 235657 at 14204 1106317 235668 at 381140 PRDM1 1106323 235674 at 442690 1106394 235750 at 126932 1106401 235758 at 11849 MGC15827 1106415 235774 at 169071 1117517 235816 s at 148656 Rgr 1106478 235843 at 119898 1106522 235890 at 31903 213 1106589 235965 at 22627 MISTI 1106722 236109 at 150458 FLJ14494 1106781 236172 at 445013 LTB4R 1106855 236255 at 455101 KAA1909 1117555 236295 s at 128357 NOD3 1106908 236313 at 72901 CDKN2B 1106935 236341 at 247824 CTLA4 1106990 236401 at 369561 1107044 236458 at 163426 1107076 .236491 at 283672 BCL2LIO 1107124 236543 at 130203 1107190 236614 at 50601 MGC10986 1107197 236621 at 40838 1107329 236761 at 439124 LHFPL3 1107348 236782 at 440508 SAMD3 1107369 236805 at 512466 1107457 236901 at 120330 ADAMTS2 1117599 236918 s at 120277 MGC27085 1107527 236981 at 14706 1107575 237033 at 424589 MGC52498 1107637 237104 at 1107762 237244 at 58597 1107838 237322 at 355618 1117644 237451 x at 34174 1107997 237493 at 126891 IL22RA2 1108088 237591 at 441601 1108200 237710 at 156135 1108237 237753 at 126232 1108323 237849 at 526982 1108347 237880 at 121476 1108467 238018 at 346333 LOC285016 1108473 238025 at 119878 FLJ34389 1108515 238071 at 98132 LCN6 1108745 238323 at 528776 TEAD2 1117747 238365_s at 158272 1108776 238376 at 513346 1108910 238536 at 351848 1108925 238552 at 136102 KIAA0853 1108961 238593 at 292088 FLJ22531 1108970 238604 at 140489 1108988 238624 at 3532 NLK 1117800 238701 x at 125166 1109058 238706 at 220277 FLJ38499 1109107 238759 at 292925 KIAA1212 1109188 238846 at 204044 TNFRSF11A 1109195 238853 at 416155 1109210 238870 at 117010 KCNK9 1109220 238880 at 445977 GTF3A 1109505 239186 at 8162 MGC39372 1109519 239201 at 348711 ALS2CR7 1117835 239205 s at 89688 CRIL 1109530 239214 at 123244 1109545 239231 at 63187 1109557 239243 at 444548 NP220 2 14 1109560 239246 at 207428 FARP1 1109603 239292 at 1109732 239427 at 374124 1109756 239453 at 530304 1117853 239479 x at 268724 1109827 239533 at 127196 GPR155 1109913 239629 at 355724 CFLAR 1110019 239744 at 1110070 239803 at 1110099 239835 at 116665 TA-KRP 1110198 239946 at 189046 1110214 239964 at 144519 TCL6 1110223 239973 at 212709 1110284 240038 at 192221 ELL2 1110309 240066 at 105623 1110313 240070 at 421750 FLJ39873 1110486 240260 at 445054 1110608 240392 at 306227 CARD14 1110610 240394 at 436906 1110740 240538 at 416810 1110852 240661 at 196026 1110871 240681 at 431753 1117977 240854 x at 1111070 240899 at 202201 1111478 241357 at 133017 ERK8 1111486 241365 at 33024 1111494 241373 at 75432 IMPDH2 1111503 241383 at 502910 KBRAS2 1111694 241592 at 157302 1111807 241751 at 6483 OFD1 1111946 241928 at 280881 1112019 242013 at 196484 1118148 242020 s at 302123 ZBP1 1112052 242052 at 525361 1112061 242064 at 43410 1112256 242293 at 143198 ING3 1112344 242406 at 163242 1118228 242520 s at 173679 1112510 242595 at 314432 C1 4orf2O 1112521 242611 at 244818 1112552 242650 at 89029 1112674 242794 at 310320 MAML3 1112689 242814 at 104879 SERPINB9 1118286 242866 x at 147381 1112762 242901 at 208179 1112764 242903 at 180866 IFNGR1 1112837 242994 at 4099 NRD1 1112849 243006 at 208965 1112871 243030 at 269493 1112935 243099 at 436677 NFAMI 1112981 243154 at 86650 1113020 243198 at 373484 LdC161577 1118347 243366 s at 528404 ITGA4 1113263 243467 at 435736 2 15 1113435 243659 at 100636 1113488 243717 at 129435 1113500 243729 at 165900 1113545 243780 at 435736 1113555 243791 at 291993 1113589 243829 at 162967 BRAF 1118414 243968 x at 415473 FCRHI 1113730 243993 at 293771 1113769 244035 at 46996 1113783 244052 at 71616 FLJ14431 1113930 244214 at 24725 MGC35521 1113972 244261 at 386334 IL28RA 1113993 244286 at 131811 1114017 244313 at 133255 1114064 244364 at 148228 MYO3A 1114109 244413 at 203041 DCALI 1114162 244467 at 526942 1114351 244677 at 445534 PERI 1114503 244845 at 170577 1114543 244887 at 156189 1118612 32625 at 438864 1118621 33307 at 239934 1130354 33323 r at 184510 SFN 1118659 35617 at 150136 1118681 36711 at 460889 1118684 36830 at 68583 MIPEP 1118708 37408 at 7835 MRC2 1118736 38340 at 96731 HIP1R 1118772 40420 at 16134 STKIO 1130378 44783 s at 234434 HEY1 1118835 47069 at 102336 ARHGAP8 1118861 49878 at 100915 PEX16 1130387 50314 i at 274422 C20orf27 1130393 58780 s at 22451 FLJ10357 1118939 60528 at 198161 PLA2G4B 1118573 632 at 435970 GSK3A 1118949 64064 at 412331 IAN4L1 1118963 65472 at 370214 1130400 74694 s at 170253 FRA 1140788 AFFX-DapX-3 at 1140834 AFFX-HSAC07/X00351 3 at 426930 ACTB 1140835 AFFX-HSAC07/X00351 5 at 426930 ACTB 1140836 AFFX-HSAC07/X00351 M at 426930 ACTB 1140842 AFFX- 169476 GAPD HUMGAPDH/M33197 3 at 1140843 AFFX- 169476 GAPD HUMGAPDH/M33197 5 at 1140844 AFFX- 169476 GAPD HUMGAPDH/M33197 M at 1140845 AFFX- 21486 STATI HUMISGF3A/M97935 3 at 1140846 AFFX- 21486 STAT1 HUMISGF3A/M97935 5 at 1140847 AFFX- 21486 STAT1 HUMISGF3A/M97935 MA at 2 16 1140848 AFFX- 21486 STATI HUMISGF3A/M97935 MB at 1140837 AFFX-HUMRGE/M10098 3 at 1140838 AFFX-HUMRGE/M10098 5 at 1140839 AFFX-HUMRGE/M10098 M at 1140791 AFFX-LysX-3 at 1140792 AFFX-LysX-5 at 1140793 AFFX-LysX-M at 1140806 AFFX-M27830 3 at 1140807 AFFX-M278305 at 1140808 AFFX-M27830 M at 1140794 AFFX-PheX-3 at 1140795 AFFX-PheX-5 at 1140796 AFFX-PheX-M at 1140797 AFFX-ThrX-3 at 1140798 AFFX-ThrX-5"at 1140799 AFFX-ThrX-M at 1140802 AFFX-TrpnX-3 at 1140803 AFFX-TrpnX-5 at 1140804 AFFX-TrpnX-M at 1140805 AFFX-hum alu at 1140809 AFFX-r2-Bs-dap-3_at 1140810 AFFX-r2-Bs-dap-5 at 1140811 AFFX-r2-Bs-dap-M at 1140812 AFFX-r2-Bs-lys-3 at 1140813 AFFX-r2-Bs-lys-5 at 1140814 AFFX-r2-Bs-lys-M at 1140815 AFFX-r2-Bs-phe-3 at 1140816 AFFX-r2-Bs-phe-5 at 1140817 AFFX-r2-Bs-phe-Mat 1140827 AFFX-r2-Bs-thr-3 s at 1140828 AFFX-r2-Bs-thr-5 s at 1140829 AFFX-r2-Bs-thr-M s at 1140820 AFFX-r2-Ec-bioB-3 at 1140821 AFFX-r2-Ec-bioB-5 at 1140822 AFFX-r2-Ec-bioB-M at 1140823 AFFX-r2-Ec-bioC-3 at 1140824 AFFX-r2-Ec-bioC-5 at 1140825 AFFX-r2-Ec-bioD-3 at 1140826 AFFX-r2-Ec-bioD-5 at 1140818 AFFX-r2-P1-cre-3 at 1140819 AFFX-r2-P1-cre-5 at 1529284 Lymph Dx 001_at 409515 MCC 1529285 Lymph Dx 002 at 348929 KIAA1219 1529286 Lymph Dx 003 at 167700 MADH5 1529287 Lymph Dx 004 s at 212787 KIAA0303 1529288 LymphDx 005 at 13291 CCNG2 1529443 Lymph Dx 006 at 88886 1529289 Lymph Dx 007 at 96557 -W4 -B 1529290 Lymph Dx 008 at 101761 N4BP3 1529291 LymphDx 009 at 104450 1529292 Lymph.Dx 010 at 1529293 Lymph Dx 011 at 113117 1529294 LymphDx 011 s at 113117 217 1529295 Lymph-Dx 012 at 116441 1529296 LymphDx 013 at 122428 1529444 Lymph-Dx 014 at 126905 1529297 Lymph Dx 015 at 132335 1529298 Lymph Dx 016 at 136707 1529299 Lymph -Dx 017 at 444290 1529300 LymphDx 018 at 449608 1529301 Lymph Dx 019 at______ 1529445 LymphDx 020 at ______________ 1529302 LymphDx 021_at 67928 ELF3 1529303 Lymph-Dx 022 at 1529304 LymphDx 022 s at 1529305 LymphDx 023 at 173957 1529306 LymphDx 024 at 190043 MGC26706 1529446 Lymph-Dx_025 at 190626 1529307 LymphDx 026 at 435736 1529308 LymphDx 027 x at 193014 1529309 LymphDx 028 at 512797 HSH2 1529310 Lymph Dx 029 x at 3136 PRKAG1 1529311 LymphDx 030 at 251214 1529312 LymphDx 031 s at 255809 1529313 Lymph-Dx 032 at 271998 1529314 Lymph Dx 033 at 1529315 Lymph-Dx 034 at 530912 1529316 Lymph-Dx 035 at 315241 ZNFI98 1529447 LymphDx 036 at 291886 1529317 Lymph Dx 037 at 1529318 Lymph Dx 038 at 291954 1529319 Lymph-Dx 039 at 103329 KIAA0970 1529320 Lymph-Dx 040 at 309149 1529321 LymphDx 041_s at 411311 IL24 1529322 LymphDx 042 x at 514291 1529323 Lymph Dx 043 at 1529324 Lymph Dx 044 at 348264 GZMH 1529325 Lymph-Dx 045 at 1529326 LymphDx 046_s at 200063 HDAC7A 1529327 LymphDx 047 s at 288986 SMN2 1529328 Lymph Ox 048 s at 369056 1529448 Lymph-Dx 049 at 369101 1529329 LymphDx 049 s at 369101 1529330 Lymph-Dx 050 at 259625 LOC153684 1529331 LymphDx 051 s at 374126 1529332 Lymph-Dx 052 at 140443 LOC134492 1529333 LymphDx 053 at 378849 1529334 Lymph Dx 054 at 529494 1529335 LymphDx 055 s at 400872 1529336 LymphDx 056 at 405474 PTK2B 1529337 Lymph-Dx 057 at 201864 06orf166 1529338 Lymph-Dx 058 s at 284275 PAK2 1529339 LymphDx 059 s at 1529449 LymphDx 060 s at 1529340 Lymph-Dx 061 at 1529341 LymphDx 062 at 153563 LY75 1529342 LymphDx 063 at 2 18 1529343 Lymph Dx 064 at 521948 1529344 LymphDx 065 at 317970 SERPINA11 1529450 Lymph.Dx 066 at 1529345 Lymph Dx 067 s at 443475 1529346 Lymph.Dx 068 at 443935 1529347 Lymph Dx 069 at 444019 1529348 Lymph Dx 070 s at 326392 SOS1 1529349 Lymph Dx 071 at 445500 1529451 Lymph Dx 072 at 396853 JMY 1529350 Lymph Dx 073 at 445884 1529351 Lymph Dx_ 074 s at 445898 1529352 Lymph Dx 075 at 446195 1529353 Lymph Dx 076 at 446198 1529354 Lymph Dx 077 at 314623 BCL11A 1529452 Lymph Dx 078 at 422550 AIMI 1529355 Lymph Dx 079 at 370675 1529356 Lymph Dx 080 at 303775 C14orf170 1529357 Lymph Dx 081 at 444651 1529358 Lymph Dx 082 at 127178 1529359 Lymph Dx 083 at 1529360 Lymph Dx 084 at 443036 1529453 Lymph Dx 085 at 372679 FCGR3A 1529361 Lymph Dx 086 s at 388681 HDAC3 1529362 Lymph Dx 087 at 329989 PLK1 1529363 Lymph Dx 088 at 311559 NOTCHI 1529364 Lymph Dx 089 at 526394 ATM 1529365 Lymph Dx 090 at 344088 TNFRSF13C 1529366 Lymph Dx 091 at 1529367 Lymph Dx 092 at 1529368 Lymph Dx 093 at 1529369 Lymph Dx 095 at 1529370 Lymph Dx 096 at 1529371 Lymph Dx 097 at 1529372 Lymph Dx 098 at 1529373 Lymph Dx 099 at 1529374 Lymph Dx 100 at 1529454 Lymph Dx 101 at 1529375 Lymph Dx 102 at 1529376 Lymph Dx 103 at 1529377 Lymph Dx 104 at 1529378 Lymph Dx 105 at 1529455 Lymph Dx 107 at 1529379 Lymph Dx 108 at 1529380 Lymph Dx 109 at 1529381 Lymph Dx 110 at 1529382 Lymph Dx 111 at 371468 CCND1 1529383 Lymph Dx 112 at 371468 CCNDI 1529456 Lymph Dx 113 at 371468 CCND1 1529384 Lymph Dx 114 at 371468 CCND1 1529385 Lymph Dx 115 at 371468 CCND1 1529386 Lymph Dx 116 at 371468 CCNDI 1529387 Lymph Dx 117 at 79241 BCL2 1529388 Lymph .Dx 118 at 79241 BCL2 1529389 Lymph Dx 119 at 79241 BCL2 2 19 1529390 Lymph Dx 120 at 79241 BCL2 1529391 Lymph Dx 121 at 79241 BCL2 1529392 Lymph Dx 122 at 352338 ACVR1C 1529393 Lymph Dx 123 s at 182081 KIAAI811 1529394 Lymph Dx 124 s at 339846 LOC91807 1529395 Lymph Dx 125 at 403201 1529396 Lymph_ Dx 126 at 512897 MGC33182 1529397 Lymph Dx 127 s at 406557 CLK4 1529398 Lymph Dx 128 at 293590 HSMDPKIN 1529399 Lymph Dx 129 at 256916 LOC203806 1529457 Lymph Dx 130 at 351818 GRK7 1529400 LymphDx 131 s at 210697 HIPK4 1529401 LymphDx 132 at 399752 MAP4K3 1529402 Lymph Dx 133 at 375836 KSR2 1529403 Lymph Dx 134 s at 511780 LMTK3 1529404 Lymph Dx 135 at 170610 MAP3K1 1529405 LymphDx,136 s at 227489 SAST 1529406 Lymph Dx 137 s at 409066 MYO3B 1529458 Lymph Dx_ 138 at 448468 NEK8 1529407 LymphDx 139 s at 284275 PAK2 1529408 Lymph Dx 141 at 336929 PSKH2 1529409 LymphDx 142 s at 351173 FLJ25006 1529410 LymphDx 143 s at 380991 SNF1LK. 1529411 Lymph Dx 144 at 80181 APEG1 1529459 Lymph Dx 145 at 411061 SRMS 1529412 Lymph Dx 146 at 512763 STK22C 1529413 Lymph Dx 147 at 232116 PRKWNK2 1529414 Lymph Dx 148 s at 352370 MGC22688 1529415 LymphDx_149_at 369523 DKFZp686A 17109 1529416 Lymph.Dx 150 s at 421349 CDKN2A 1529417 Lymph Dx 151 at 421349 CDKN2A 1529418 Lymph Dx 152 at 421349 CDKN2A 1529419 LymphDx 153 s at 104182 1529420 LymphDx 154 at 272295 IL17F, 1529421 Lymph Dx 156 at 375043 IL27 1529422 Lymph Dx 157 s at 375184 IL23R 1529423 LymphDx 158 at 381264 ITGAD 1529424 LymphDx 159 s at 512683 CCL3L1 1529425 Lymph Dx 160 at 406228 IL9R 1529426 Lymph Dx 162 at 406744 IL28B 1529427 Lymph Dx 163 at 406745 IL29 1529428 Lymph Dx 164 at 415768 NGFR 1529429 Lymph Dx 165 at 434103 IL17D 1529430 Lymph Dx 166 at 444484 SPHK2 1529431 Lymph Dx 167 at 1529432 Lymph Dx 168 at 1529433 Lymph.Dx 168 x at 1529434 LynphDx 171,at 103995 FLJ27099 1529435 Lymph Dx 172 s at 1529436 Lymph Dx 174 at 1529437 Lymph Dx 175 at 445162 BTLA 220 REFERENCES 1. Alizadeh, A.A., et al. 1998. Probing lymphocyte biology by genomic scale gene expression analysis. J Clin Immunol 18:373-79. 2. Alizadeh, A.A., et al. 1999. The Lymphochip: a specialized cDNA 5 microarray for the genomic-scale analysis of gene expression in normal and malignant lymphocytes. Cold Spring Harbor Symp Quant Biol 64:71-78. 3. Alizadeh, A.A., et al. 2000. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403:503-511. 4. Alon, U., et al. 1999. Broad patterns of gene expression revealed by 0 clustering analysis of tumor and normal colon tissues probed by oligonucleotide arrays. Proc Nati Acad Sci USA 96:6745-6750. 5. Bayes, T. 1763. An essay towards solving a problem in the doctrine of chances. Phil Trans Roy Soc London 53:370. 6. Chee, M., et al. 1996. Accessing genetic information with high density 5 DNA arrays. Science 274:610-14. 7. Cho, R.J., et al. 1998. A genome-wide transcriptional analysis of the mitotic cell cycle. Mol Cell 2:65-73. 8. Chu, S., et al. 1998. The transcriptional program of sporulation in budding yeast. Science 282:699-705. 20 9. Copie-Bergman, C., et al. 2002. MAL expression in lymphoid cells: further evidence for MAL as a distinct molecular marker of primary mediastinal large B-cell lymphomas. Mod Pathol 15:1172-1180. 10. Copie-13ergman, C., et al. 2003. Interleukin 4-induced gene 1 is activated in primary mediastinal large B-cell lymphoma. Blood 101:2756-2761.

22 1 11. DeRisi, J., et al. 1996. Use of a cDNA microarray to analyze gene expression patterns in human cancer. Nat Genet 14:457-60. 12. DeRisi, J.L., lyer, V.R., Brown, P.O. 1997. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278:680-86. 5 13. Drapner, H. 1966. Applied regression. Wiley, New York. 14. Dudoit, S., Fridlyand, J., Speed, T.P. 2002. Comparison of discrimination methods for the classification of tumors using gene expression data. J Am Stat Assoc 97:77-87. 15. Eisen, M.B., Spellman, P.T., Brown, P.O., Botstein, D. 1998. Cluster 0 analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA 95:14863-14868. 16. Fisher, R.I., et al. 1993. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma. N Engl J Med 328:1002-1006. .5 17. Furey, T.S., et al. 2000. Support vector machine classification and validation of cancer tissue samples using microarray expression data. Bioinformatics 16:906-914. 18. Golub, T.R., et al. 1999. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring, Science 286:531 20 537. 19. Gress, T.M., et al. 1996. A pancreatic cancer-specific expression profile. Oncogene 13:1819-30. 20. Harris, N.L., et al. 1994. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. 25 Blood 84:1361-1392.

222 21. Heller, R.A., et al. 1997. Discovery and analysis of inflammatory disease-related genes using cDNA microarrays. Proc Nati Acad Sci USA 94:2150 55. 22. Holstege, F.C., et al. 1998. Dissecting the regulatory circuitry of a 5 eukaryotic genome. Cell 95:717-728. 23. Irizarry, R.A., et al. 2003. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4:249-264. 24. Hills, M. 1966. Allocation rules and error rates. J Royal Statis Soc Series B 28:1-31. 0 25. Jaffe, E.S., Harris, N.L., Stein, H., Vardiman, J.W. 2001. Tumors of hematopoietic and lymphoid tissues. IARC Press, Lyon. 26. Khouri, I.F., et al. 1998. Hyper-CVAD and high-dose methotrexate/cytarabine followed by stem-cell transplantation: an active regimen for aggressive mantle-cell lymphoma. J Clin Oncol 12:3803-3809. 5 27. Kohonen, T. 1997. Self-organizing maps. Springer Press, Berlin. 28. Lashkari, D.A., et al. 1997. Yeast microarrays for genome wide parallel genetic and gene expression analysis. Proc Natl Acad Sci USA 94:13057-62. 29. Li, C., Wong, W.H. 2001. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Nat Acad Sci USA 20 98:31-36. 30. Lipshutz, R.J., et al. 1995. Using oligonucleotide probe arrays to access genetic diversity. Biotechniques 19:442-47. 31. Lockhart, D.J., et al. 1996. Expression monitoring by hybridization to high-density oligonucleotide arrays. Nat Biotechnol 14:1675-80.

223 32. Pease, A.C., et al. 1994. Light generated oligonucleotide arrays for rapid DNA sequence analysis. Proc NatI Acad Sci USA 91:5022-26. 33. Pietu, G., et al. 1996. Novel gene transcripts preferentially expressed in human muscles revealed by quantitative hybridization of a high density cDNA 5 array. Genome Res 6:492-503. 34. Radmacher, M.D., McShane, L.M., Simon, R. 2002. A paradigm for class prediction using gene expression profiles. J Comput Biol 9:505-511. 35. Ramaswamy, S., et al. 2001. Multiclass cancer diagnosis using tumor gene expression signatures. Proc Natl Acad Sci USA 98:15149-15154. 10 36. Ransohoff, D.F. 2004. Rules of evidence for cancer molecular-marker discovery and validation. Nat Rev Cancer 4:309-314. 37. Rosenwald, A., et al. 2002. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. New Engl J Med 346:1937-1947. 15 38. Rosenwald, A., et al. 2003. The proliferation gene expression signature is a quantitative integrator of oncogenic events that predicts survival in mantle cell lymphoma. Cancer Cell 3:185-197. 39. Schena, M., Shalon, D., Davis, R.W., Brown, P.O. 1995. Quantitative monitoring of gene expression pattems with a complementary DNA microarray. 20 Science 270:467-70. 40. Schena, M., et al. 1996. Parallel human genome analysis: microarray based expression monitoring of 1000 genes. Proc NatI Acad Sci USA 93:10614-19. 41. Shaffer, A.L., et al. 2001. Signatures of the immune response. Immunity 15:375-385.

224 42. Shalon, D., Smith, S.J., Brown, P.O. 1996. A DNA microarray system for analyzing complex DNA samples using two-color fluorescent probe hybridization. Genome Res 6:639-45. 43. Shipp, M.A., et al. 2002. Diffuse large B-cell lymphoma outcome 5 prediction by gene-expression profiling and supervised machine learning. Nat Med 8: 68-74. 44. Southern, E.M., Maskos, U., Elder, J.K. 1992. Analyzing and comparing nucleic acid sequences by hybridization to arrays of oligonucleotides: evaluation using experimental models. Genomics 13:1008-17. 0 45. Southern, E.M., et al. 1994. Arrays of complementary oligonucleotides for analysing the hybridisation behaviour of nucleic acids. Nuct Acids Res 22:1368 73. 46. Spellman, P.T., et al. 1998. Comprehensive identification of cell cycle regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. 5 Mol Biol Cell 9:3273-3297. 47. Tamayo, P., et al. 1999. Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation. Proc Natl Acad Sci USA 96:2907-2912. 48. Tavazoie, S., et al. 1999. Systematic determination of genetic network 20 architecture. Nat Genet 22:281-285. 49. Tibshirani, R., Hastie, T., Narasimhan, B., Chu, G. 2002. Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc Natl Acad Sci USA 99:6567-6572. 50. Velculescu, V.E., Zhang, L., Vogelstein, B., Kinzler, K.W. 1995. Serial 25 analysis of gene expression. Science 270:484-87.

225 51. Wodicka, L., et al. 1997. Genome-wide expression monitoring in Saccharomyces cerevisiae. Nat Biotechnol 15:1359-6714. 52. Wright, G., et al. 2003. A gene expression-based method to diagnose clinically distinct subgroups of diffuse large B cell lymphoma. Proc Nati Acad Sci 5 USA 100:9991-9996.

Claims (95)

1. A composition comprising the probes listed in Table 2.

2. The composition of claim 1 wherein said probes comprise a microarray.

3. A method for generating a survival prediction model for a lymphoma 5 comprising the steps of: a) obtaining one or more biopsy samples of said lymphoma, wherein said biopsy samples are obtained from subjects with known survival data; b) obtaining gene expression data for a set of genes in said one or more biopsy samples; 10 c) identifying genes with expression patterns associated with longer survival; d) identifying genes with expression patterns associated with shorter survival; e) applying hierarchical clustering to those genes identified in step (c) to identify one or more gene expression signatures; f) applying hierarchical clustering to those genes identified in step (d) to 15 identify one or more gene expression signatures; g) for each gene expression signature identified in steps (e) and (f), averaging the expression level of each gene within the gene expression signature to obtain a gene expression signature value; and h) generating a multivariate survival prediction model using the gene 20 expression signature values obtained in step (g).

4. A method for predicting survival in a follicular lymphoma (FL) subject comprising the steps of: a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; 227 c) averaging the gene expression level of genes from an immune response-1 gene expression signature to obtain an immune response-I gene expression signature value; d) averaging the gene expression level of genes from an immune response-2 5 gene expression signature to obtain an immune response-2 gene expression signature value; f) calculating a survival predictor score using an equation: [2.71*(immune response-2 gene expression signature value)] - [2.36*(immune response-1 gene expression signature value)];wherein a higher survival predictor score is associated 10 with worse survival.

5. The method of claim 4 wherein the step of obtaining gene expression data further comprises use of a microarray.

6. The method of claim 4 wherein the immune response-1 gene expression signature comprises at least one gene selected from the group consisting of (listed [5 by UNIQID): 1095985, 1096579, 1097255, 1097307, 1097329, 1097561, 1098152, 1098405, 1098548, 1098893, 1099053, 1100871, 1101004,1103303,1107713, 1115194,1119251, 1119838, 1119924, 1120196,1120267, 1121313, 1121406, 1121720,1122956,1123038,1123092,1123875,1124760,1128356,1128395, 1132104, 1133408,1134069,1134751,1134945,1135743,1135968, 1136048, 20 1136087, 1137137, 1137289, 1137534, 1139339, 1139461, 11400391,1140524, and 1140759.

7. The method of claim 4 wherein the immune response-2 gene expression signature comprises'at least one gene selected from the group consisting of (listed by UNIQID): 1118755, 1118966, 1121053, 1121267, 1121331, 1121766, 1121852, 25 1122624, 1122679, 1122770, 1123767, 1123841,1126097, 1126380, 1126628, 228 1126836, 1127277, 1127519, 1127648, 1128483, 1128818, 1129012, 1129582, 1129658, 1129705, 1129867, 1130003, 1130388, 1131837, 1133843,1133949, 1134447, 1135117, 1136017, 1137478, 1137745, 1137768, 1138476, 1138529, 1138601, 1139862, 1140189, and 1140389. 5

8. A method for predicting survival in a follicular lymphoma (FL) subject comprising the steps of: a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; c) averaging the gene expression level of genes from a B-cell differentiation 10 gene expression signature to obtain a B-cell differentiation gene expression signature value; d) averaging the gene expression level of genes from a T-cell gene expression signature to obtain a T-cell gene expression signature value; e) averaging the gene expression level of genes from a macrophage gene 15 expression signature to obtain a macrophage gene expression signature value; f) calculating a survival predictor score using an equation: [2.053*(macrophage gene expression signature value)] - [2.344*(T-cell gene expression signature value)] - [0.729*(B-cell gene expression signature value)]; wherein a higher survival predictor score is associated with worse survival. 20

9. The method of claim 8 wherein the step of obtaining gene expression data further comprises use of a microarray.

10. The method of claim 8 wherein the B-cell differentiation gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1102859, 1098862, 1099291, 1101439, 1120316, 1139017, 1130922, 229 1130923, 1119350, 1097897,1097901,1120976, 1119813, 1105935,1111070, 1114726,1108988,1096035, and 1123298.

11. The method of claim 8 wherein the T-cell gene expression signature comprises at least one gene selected from the group consisting of (listed by 5 UNIQID): 1134945, 1134069, 1137809, 1119251, 1096579, 1101004, 1137137, 1100871,1139461,1128395,1119880, 1130676,1130668, 1135968,1097329, 1098548,1123038, 1128356, 1133408, 1140524,1119838, 1097255,1098152, 1115194,1124760, 1120267,1137289,1137534, 1097307, 1123613,1121720, 1120196,1136087,1132104,1140391,1098405,1135743,1136048,1123875, 10 1098893, 1097561, 1122956, 1121406, 1125532, 1138538, 1103303, and 1119924.

12. The method of claim 8 wherein the macrophage gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1123682, 1099124, 1123401, 1134379, 1137481, 1132220, 1119400, 1131119,1123566, 1138443, 1127943, 1119998, 1132433, 1119260, and 1098278. 15

13. A method for predicting survival in a follicular lymphoma (FL) subject comprising the steps of: a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; c) averaging the gene expression level of genes from a B-cell differentiation 20 gene expression signature to obtain a B-cell differentiation gene expression signature value; d) averaging the gene expression level of genes from a T-cell gene expression signature to obtain a T-cell gene expression signature value; e) averaging the gene expression level of genes from a macrophage gene 25 expression signature to obtain a macrophage gene expression signature value; 230 f) calculating a survival predictor score using an equation: [1.51*(macrophage gene expression signature value)] - [2.1 1*(T-cell gene expression signature value)] - [0.505*(B-cell differentiation gene expression signature value)]; wherein a higher survival predictor score is associated with worse survival. 5

14. The method of claim 13 wherein the step of obtaining gene expression data further comprises use of a microarray.

15. The method of claim 13 wherein the B-cell differentiation gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1102859, 1098862, 1099291, 1101439, 1120316, 1139017, 1130922, 10 1130923,1119350,1097897,1097901,1120976,1119813,1105935,1111070, 1114726,1108988,1096035, and 1123298.

16. The method of claim 13 wherein the T-cell gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQD): 1134945, 1134069, 1137809, 1119251, 1096579, 1101004, 1137137, 15 1100871,1139461,1128395,1119880,1130676,1130668,1135968,1097329, 1098548,1123038,1128356,1133408,1140524,1119838,1097255,1098152, 1115194,1124760,1120267,1137289,1137534,1097307,1123613, 1121720, 1120196,1136087,1132104,1140391,1098405,1135743,1136048,1123875, 1098893, 1097561, 1122956,1121406,1125532,1138538,1103303, and 1119924. 20

17. The method of claim 13 wherein the macrophage gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1123682,1099124,1123401,1134379,1137481, 1132220, 1119400, 1131119, 1123566, 1138443, 1127943, 1119998, 1132433, 1119260, and 1098278.

18. A method for predicting survival in a diffuse large B cell lymphoma (DLBCL) 25 suibiect comorisina the steos of: 23 1 a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; c) averaging the gene expression level of genes from an ABC DLBCL high gene expression signature to obtain an ABC DLBCL high gene expression signature 5 value; d) averaging the gene expression level of genes from a lymph node gene expression signature to obtain a lymph node gene expression signature value; e) averaging the gene expression level of genes from an-MHC class Il gene expression signature to obtain an MHC class 11 gene expression signature value; 10 f) calculating a survival predictor score using an equation: [0.586*(ABC DLBCL high gene expression signature value)],- [0.468*(lymph node gene expression signature value)] - [0.336*(MHC class Il gene expression signature value)];wherein a higher survival predictor score is associated with worse survival.

19. The method of claim 18 wherein the step of obtaining gene expression data 15 comprises use of a microarray.

20. The method of claim 18 wherein the ABC DLBCL high gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1134271, 1121564, 1119889, 1133300, 1106030, 1139301, 1122131, 1114824, 1100161, and 1120129. 20

21. The method of claim 18 wherein the lymph node gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1097126, 1120880, 1098898, 1123376, 1128945, 1130994, 1124429, 1099358, 1130509, 1095985, 1123038, 1133700, 1122101, and 1124296. 232

22. The method of claim 18 wherein the MHC class I gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQD): 1123127, 1136777, 1137771, 1134281, 1136573, and 1132710.

23. A method for predicting survival in a diffuse large B cell lymphoma (DLBCL) 5 subject comprising the steps of: a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; c) averaging the gene expression level of genes from a lymph node gene expression signature to obtain a lymph node gene expression signature value; 10 d) averaging the gene expression level of genes from a germinal B cell gene expression signature to obtain a germinal B cell gene expression signature value; e) averaging the gene expression level of genes from a proliferation gene expression signature to obtain a proliferation gene expression signature value; f) averaging the gene expression level of genes from an MHC class 11 gene 15 expression signature to obtain a proliferation gene expression signature value; g) calculating a survival predictor score using an equation: [-0.4337*(lymph node gene expression signature)] + [0.09*(proliferation gene expression signature)] - [0.4144*(germinal center B-cell gene expression signature)] - [0.2006*(MHC class Il gene expression signature)]; 20 wherein a higher survival predictor score is associated with worse survival.

24. The method of claim 23 wherein the step of obtaining gene expression data comprises use of a microarray.

25. The method of claim 23 wherein the lymph node gene expression signature comprises at least one gene selected from the group consisting of (listed by 233 UNIQID): 1097126,1099028,1099358,1101478,1103497, 1121029, 1124429, 1135068,1136051, and 1136172.

26. The method of claim 23 wherein the proliferation gene expression signature comprises at least one gene selected from the group consisting of (listed by 5 UNIQID): 1096903, 1120583, 1123289, 1131808, 1133102, and 1136595.

27. The method of claim 23 wherein the germinal center B-cell gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1099686, 1099711, 1103390, 1106025, 1128287, 1132520, 1138192, 1529318,1529344, 1529352, 1096570, 1097897, 1097901, 1098611, 1100581, 10 1115034,1120090, 1120946,1121248,1123105,1125456,1128694,1128787, 1132122,1136269, 1136702, 1139230, 1529292, and 1529295.

28. The method of claim 23 wherein the MHC class I gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQD): 1136777 and 1136877. 15

29. A method for predicting survival in a diffuse large B cell lymphoma (DLBCL) subject comprising the steps of: a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; c) averaging the gene expression level of genes from a lymph node gene 20 expression signature to obtain a lymph node gene expression signature value; d) averaging the gene expression level of genes from a germinal B cell gene expression signature to obtain a germinal B cell gene expression signature value; e) averaging the gene expression level of genes from an MHC class Il gene expression signature to obtain a proliferation gene expression signature value; 234 f) calculating a survival predictor score using an equation: [-0.32*(lymph node gene expression signature)] - [0.176*(germinal B cell gene expression signature)] [0.206*(MHC class Il gene expression signature)]; wherein a higher survival predictor score is associated with worse survival. 5

30. The method of claim 29 wherein the step of obtaining gene expression data further comprises use of a microarray.

31. The method of claim 29 wherein the step of obtaining gene expression data further comprises RT-PCR.

32. The method of claim 29 wherein the lymph node gene expression signature 10 comprises at least one gene selected from the group consisting of (listed by UNIQID): 1097126, 1099358, and 1121029.

33. The method of claim 29 wherein the germinal center B-cell gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1099686, 1529318, and 1529344. 15

34. The method of claim 29 wherein the MHC class Il gene expression signature comprises at least one gene selected from the group consisting of (listed by UNIQID): 1136777 and 1136877.

35. A method for predicting survival in a mantle cell lymphoma (MCL) subject comprising the steps of: 20 a) obtaining a biopsy sample from said subject; b) obtaining gene expression data for a set of genes in said biopsy sample; c) averaging the gene expression level of genes from a proliferation gene expression signature to obtain a proliferation gene expression signature value; f) calculating a survival predictor score using an equation: [1.66*(proliferation 25 gene expression signature value)]; 235 wherein a higher survival predictor score is associated with worse survival.

36. The method of claim 35 wherein the step of obtaining gene expression data further comprises use of a microarray.

37. The method of claim 35 wherein the proliferation gene expression signature 5 comprises at least one gene selected from the group consisting of (listed by UNIQD): 1097290,1101295,1119729, 1120153, 1120494,1124745, 1126148, 1130618, 1134753, 1139654, and 1140632.

38. The method of claim 35 wherein the proliferation gene expression signature comprises at least one gene selected from the group consisting of (listed by 10 UNIQID): 1119294,1119729, 1120153, 1121276, 1123358, 1124178, 1124563, 1130799, 1131274, 1131778, 1132449, 1135229, and 1136585.

39. A method for determining the probability that a sample X belongs to a first lymphoma type or a second lymphoma type comprising the steps of: a) identifying a set of genes G that are differentially expressed between a first 15 lymphoma type and a second lymphoma type; b) calculating a series of scale factors, wherein each scale factor represents a difference in gene expression between said first lymphoma type and said second lymphoma type for one of the genes identified in step (a); c) generating a series of linear predictor scores for a set of known samples 20 belonging to said first lymphoma type and a set of known samples belonging to said second lymphoma type based on the expression of the genes identified in step (a); d) obtaining gene expression data for the genes identified in step (a) for sample X; 236 e) generating a linear predictor score for sample X based on the expression of the genes identified in step (a); f) calculating a probability q that sample X belongs to said first lymphoma type by: 5 q = #(LPS(X); A,,4,) #(LPS(X); A,&,) + #(LPS(X); A 2 , 62) wherein LPS(X) is the linear predictor score for sample X, #(x; u, o-) is the normal density function with mean p and standard deviation a-, A and 6, are the mean and variance of the linear predictor scores for said known samples belonging to said first lymphoma type, and A 2 and6 2 are the mean and variance of the linear predictor LO scores for said known samples belonging to said second lymphoma type.

40. The method of claim 39 wherein the linear predictor scores are calculated by: LPS(S) = Z tjSj, jEG wherein S; is the expression of gene j in a sample S and t; is the scale factor representing the difference in expression of gene j between said first lymphoma type 15 and said second lymphoma type.

41. The method of claim 39 wherein said scale factors are t-statistics.

42. The method of claim 39 wherein said first lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma 20 (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, iymphoplasmacytic lymphoma, post-transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB 237 DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and primary mediastinal B cell lymphoma (PMBL).

43. The method of claim 39 wherein said second lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell 5 lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, lymphoplasmacytic lymphoma, post-transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB 10 DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and primary mediastinal B cell lymphoma (PMBL).

44. The method of claim 39 wherein said sample is classified as said first lymphoma type if said probability q is greater than 90%.

45. The method of claim 39 wherein said set of genes G excludes genes 15 belonging to a proliferation gene expression signature and genes belonging to a lymph node gene expression signature.

46. The method of claim 39 wherein step (d) further comprises use of a microarray.

47. A method for determining the lymphoma type of a sample X comprising the 20 steps of: a) identifying a set of genes G that are differentially expressed between a first lymphoma type and a second lymphoma type; 238 b) calculating a series of scale factors, wherein each scale factor represents a difference in gene expression between said first lymphoma type and said second lymphoma type for one of the genes identified in step (a); c) generating a series of linear predictor scores for a set of known samples 5 belonging to said first lymphoma type and a set of known samples belonging to said second lymphoma type based on the expression of the genes identified in step (a); d) obtaining gene expression data for the genes identified in step (a) for sample X; e) generating a linear predictor score for sample X based on the expression 10 of the genes identified in step (a); f) calculating a probability q that sample X belongs to said first lymphoma type by: q =$(LPS(X); A,,&.) $(LPS(X); A 1 ,6 1 ) + #(LPS(X); A12,62) wherein LPS(X) is the linear predictor score for sample X, #(x; p, a) is the normal 15 density function with mean p and standard deviation a-, A, and 6, are the mean and variance of the linear predictor scores for said known samples belonging to the first lymphoma type, and A2 and& 2 are the mean and variance of the linear predictor scores for said known samples belonging to the second lymphoma type; g) repeating steps (a)-(f) with other lymphoma types replacing the second 20 lymphoma type; h) repeating steps (a)-(g) with other lymphoma types replacing the first lymphoma type.

48. The method of claim 47 wherein the linear predictor scores are calculated by: 239 LPS(S).= EtjSj, jeG wherein Sj is the expression of gene jin a sample S and t is the scale factor representing the difference in expression of gene j between said first lymphoma type and said second lymphoma type. 5

49. The method of claim 47 wherein the scale factors are t-statistics.

50. The method of claim 47 wherein the first lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, 10 multiple myeloma, lymphoplasmacytic lymphoma, post-transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and primary mediastinal B cell lymphoma (PMBL). 15

51. The method of claim 47 wherein the second lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, lymphoplasmacytic lymphoma, post-transplant 20 lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and primary mediastinal B cell lymphoma (PMBL). 240

52. The method of claim 47 wherein the first lymphoma type is mantle cell lymphoma (MCL).

53. The method of claim 52 wherein the second lymphoma type is activated B cell-like diffuse large B cell lymphoma (ABC DLBCL). 5

54. The method of claim 53 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1103711, 1133111, 1137987, 1132835, 1109505, 1139054, 1119361, 1115226, 1101211, 1118963, 1096503, 1127849,1099204, 1098840, 1139444, 1106855, 1126695, 1120137, 1133011, and 1133192. 10

55. The method of claim 52 wherein the second lymphoma type is Burkitt lymphoma (BL).

56. The method of claim 55 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1120900, 1112061, 1109505, 1133099, 1106855, 1110070, 1121739, 1098840, 1132833, 1121693, 15 1123760, 1125964, 1112306, 1096070, 1129943, 1118749, 1098954, 1134749, 1131860, and 1123148.

57. The method of claim 52 wherein the second lymphoma type is follicular hyperplasia (FH).

58. The method of claim 57 wherein the differentially expressed genes are 20 selected from the group consisting of (listed by UNIQID): 1132834, 1100873, 1109603,1139411, 1106855, 1125193, 1137450, 1100258, 1133167, 1136831, . 1138222, 1099437, 1140236, 1114109, 1098277, 1135138, 1103304, 1128460, 1121953, and 1129281.

59. The method of claim 52 wherein the second lymphoma type is follicular 25 lymphoma (FL). 24 1

60. The method of claim 59 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1132835, 1096070, 1103711, 1137987, 1109505, 1098840, 1130926, 1096396, 1132734, 1139393, 1115537, 1102215, 1124585, 1137561, 1100581, 1124646, 1114543, 1120090, 5 1123731, and 1133192.

61. The method of claim 52 wherein the second lymphoma type is germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL).

62. The method of claim 61 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1098840, 1132835, 10 1137987, 1098954, 1103711, 1096070, 1139393, 1127849, 1098156,1128845, 1129943, 1140116, 1106855, 1120900, 1127371, 1119361, 1120854, 1098277, 1140127, and 1100581.

63. The method of claim 52 wherein the second lymphoma type is mucosa associated lymphoid tissue lymphoma (MALT). 15

64. The method of claim 63 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1132834, 1101987, 1100873, 1130764, 1102178, 1098277, 1130926, 1098694, 1103711, 1138099, 1120854, 1102215, 1121739, 1096070, 1101211, 1120825, 1099437, 1096503, 1135927, and 1120645. 20

65. The method of claim 52 wherein the second lymphoma type is primary mediastinal B cell lymphoma (PMBL).

66. The method of claim 65 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1132834, 1100873, 1096503, 1098840, 1124734, 1135102, 1103711, 1140416, 1121757, 1140236, 242 1099140,1099549,1139054,1138818,1109444,1124534, 1098277, 1131687, 1125112, and 1125397.

67. The method of claim 52 wherein the second lymphoma type is post-transplant lymphoproliferative disorder (PTLD). 5

68. The method of claim 67 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1109603, 1138222, 1135138,1134230,1139411, 1140416, 1132834,1121739,1098156,1099270, 1139012,1120854, 1120985,1115952,1120825,1131636, 1136706,1113560, 1133851, and 1137459. 10

69. The method of claim 52 wherein the second lymphoma type is small cell lymphocytic lymphoma (SLL).

70. The method of claim 69 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 1132834, 1101987, 1103711, 1096070, 1130926, 1120645, 1138099, 1097887, 1099941,1130373, 15 1110957,1130320,1124373,1128813,1131130,1120825,1119752,1131854, 1105801, and 1097824.

71. The method of claim 52 wherein the second lymphoma type is splenic lymphoma.

72. The method of claim 71 wherein the differentially expressed genes are 20 selected from the group consisting of (listed by UNIQID): 1106855, 1121739, 1111850,1098024, 1130764, 1135342,1097218,1117193, 1139564,1132834, 1131130,1131756, 1102187,1098195,1101211,1136673,1139116,1098694, 1120519, and 1114916.

73. The method of claim 47 wherein the first lymphoma type is activated B cell 25 like diffuse large B cell lymphoma (ABC DLBCL). 243

74. The method of claim 73 wherein the second lymphoma type is germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL).

75. The method of claim 74 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 19375, 19346, 19227, 5 16049,32529,24729,24899,19348,27565,17227,26919,24321,29385,16858, 31801,19234,26385,24361,24570,24904,24429,28224,27673,24376,17496, 17218, and 28338.

76. The method of claim 74 wherein the differentially expressed genes are selected from the group consisting of (listed by UNIQID): 24729, 17227, 26907, 10 27565,16858,24899,16947,16049,26385,27673,24429,17218,28338,and

17496.

77. The method of claim 47 wherein said set of genes G excludes genes belonging to a proliferation gene expression signature and genes belonging to a lymph node gene expression signature. 15

78. The method of claim 47 wherein step (d) further comprises use of a microarray.

79. A method for determining the lymphoma type of a sample X comprising the steps of: a) creating a series of lymphoma type pairs, wherein each lymphoma type 20 pair represents a combination of a first lymphoma type and a second lymphoma type; b) for each lymphoma type pair, obtaining gene expression data for a set of genes G in said first lymphoma type and said second lymphoma type; 244 c) calculating a series of scale factors, wherein each scale factor represents a difference in gene expression between said first lymphoma type and said second lymphoma type for one of the genes identified in step (b); d) identifying z genes from said set of genes G with the largest scale factors; 5 e) generating a series of linear predictor scores for a set of known samples belonging to said first lymphoma type and a set of known samples belonging to said second lymphoma type based on the expression of the genes identified in step (d), wherein said series of linear predictor scores is generated using between I and z of the genes identified in step (d); 10 f) selecting a set of genes between 1 and z from step (e) that generates the largest difference in linear predictor score between said first lymphoma type and said second lymphoma type; g) obtaining gene expression data for the genes identified in step (f) for sample X; 15 h) generating a linear predictor score for sample X based on the expression of the genes selected in step (f); i) calculating a probability q that sample X belongs to said first lymphoma type by: q #(LPS(X); A,6,) O(LPS(X); A 1 ,& 1 ) + #(LPS(X); f12,&2) 20 wherein LPS(X) is the linear predictor score for sample X, #(x; p, a) is the normal density function with mean p and standard deviation o-, A and diare the mean and variance of the linear predictor scores for said set of known samples belonging to said first lymphoma type, and A2 and& 2 are the mean and variance of the linear 245 predictor scores for said known samples belonging to said second lymphoma type, and wherein a high probability q indicates that sample X belongs to said first lymphoma type, a low probability q indicates that sample X belongs to said second lymphoma type, and a middle probability q indicates that sample X belongs to 5 neither lymphoma type.

80. The method of claim 62 wherein the linear predictor scores are calculated by: LPS(S) = tS, jeG wherein Sj is the expression of gene j in a sample S and t is the scale factor representing the difference in expression of gene between said first lymphoma type [0 and said second lymphoma type.

81. The method of claim 79 wherein said scale factors are t-statistics.

82. The method of claim 79 wherein said first lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma [5 (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, lymphoplasmacytic lymphoma, post-transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and 20 primary mediastinal B cell lymphoma (PMBL).

83. The method of claim 79 wherein said second lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, 246 multiple myeloma, lymphoplasmacytic lymphoma, post-transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and 5 primary mediastinal B cell lymphoma (PMBL).

84. The method of claim 79 wherein a cut-off point between said high probability q and said middle probability q and a cut-off point between said middle probability q and said low probability q is determined by the following steps: i) ranking one or more samples of known lymphoma type according to their 10 probability q; ii) analyzing each cut-off point between adjacent samples by: 3.99 * [(% of said first lymphoma type misidentified as said second lymphoma type) + (% of said second lymphoma type misidentified as said first lymphoma type)] + of said first lymphoma type classified as belonging to neither lymphoma type) + (% 15 of said second lymphoma type classified as belonging to neither lymphoma type)], wherein the final cut-off points are those that minimize this equation.

85. The method of claim 79 wherein z = 100.

86. The method of claim 79 wherein step (g) further comprises use of a microarray. 20

87. A method for determining the lymphoma type of a sample X comprising the steps of: a) creating a series of lymphoma type pairs, wherein each lymphoma type pair represents a combination of a first lymphoma type and a second lymphoma type; 247 b) for each lymphoma type pair, obtaining gene expression data for a set of genes G in said first lymphoma type and said second lymphoma type; c) calculating a series of scale factors, wherein each scale factor represents a difference in gene expression between said first lymphoma type and said second 5 lymphoma type for one of the genes identified in step (b); d) placing each gene in said set of genes G into one of n gene-list categories, wherein placement in a gene-list category indicates correlation between expression of said gene and expression of a gene expression signature; e) within each gene-list category, identifying z genes with the largest scale 10 factors; f) generating a series of linear predictor scores for a set of known samples belonging to said first lymphoma type and a set of known samples belonging to said second lymphoma type based on the expression of the genes identified in step (e), wherein said series of linear predictor scores is generated using between 1 and z of 15 the genes identified in step (e); g) selecting a set of genes between 1 and z from step (f) that generates the largest difference in linear predictor scores between said first lymphoma type and said second lymphoma type; h) measuring expression of the set of genes selected in step (g) in sample X; 20 i) generating a linear predictor score for sample X based on the expression of the genes selected in step (f); j) calculating a probability q that sample X belongs to said first lymphoma type by: q #(LPS(X); i,d) $(LPS(X); A, 6- 1 ) +# (LPS(X); 4,2) 248 wherein LPS(X) is the linear predictor score for sample X, #(x; P, a) is the normal density function with mean u and standard deviation a-, A, and ci, are the mean and variance of the linear predictor scores for said set of known samples belonging to said first lymphoma type, and A2 and5 2 are the mean and variance of the linear 5 predictor scores for said known samples belonging to said second lymphoma type, and wherein a high probability q indicates that sample X belongs to said first lymphoma type, a low probability q indicates that sample X belongs to said second lymphoma type, and a middle probability q indicates that sample X belongs to neither lymphoma type, and wherein a cut-off point between said high probability q 10 and said middle probability q and a cut-off point between said middle probability q and said low probability q is determined by the following steps: i) ranking one or more samples of known lymphoma type according to their probability q; ii) analyzing each cut-off point between adjacent samples by: 15 3.99 * [(% of said first lymphoma type misidentified as said second lymphoma type) + (% of said second lymphoma type misidentified as said first lymphoma type)] + of said first lymphoma type classified as belonging to neither lymphoma type) + (% of said second lymphoma type classified as belonging to neither lymphoma type)], wherein the final cut-off points are those that minimize this equation. 20

88. The method of claim 87 wherein the linear predictor scores are calculated by: LPS(S)= tS , jeG 249 wherein S; is the expression of gene j in a sample S and t is the scale factor .representing the difference in expression of gene j between said first lymphoma type and said second lymphoma type.

89. The method of claim 87 wherein said scale factors are t-statistics. 5

90. The method of claim 87 wherein said first lymphoma type is selected from the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, lymphoplasmacytic lymphoma, post-transplant 10 lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and primary mediastinal B cell lymphoma (PMBL).

91. The method of claim 87 wherein said second lymphoma type is selected from 15 the group consisting of: follicular lymphoma (FL), Burkitt lymphoma (BL), mantle cell lymphoma (MCL), follicular hyperplasia (FH), small cell lymphocytic lymphoma (SLL), mucosa-associated lymphoid tissue lymphoma (MALT), splenic lymphoma, multiple myeloma, lymphoplasmacytic lymphoma, post-transplant lymphoproliferative disorder (PTLD), lymphoblastic lymphoma, nodal marginal zone 20 lymphoma (NMZ), germinal center B cell-like diffuse large B cell lymphoma (GCB DLBCL), activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) and primary mediastinal B cell lymphoma (PMBL).

92. The method of claim 87 wherein n = 3. 250

93. The method of claim 92, wherein said gene-list categories are a lymph node gene expression signature, a proliferation gene expression signature, and a standard gene expression signature, wherein said standard gene expression signature includes those genes not included in said lymph node and proliferation 5 gene expression signatures.

94. The method of claim 93, wherein step (g) further comprises generating four linear predictor scores using the set of genes selected therein, wherein: a) the first linear predictor score is generated using genes from the lymph node, proliferation, and standard gene expression signatures; [0 b) the second linear predictor score is generated using genes from the standard gene expression signature only; c) the third linear predictor score is generated using genes from the standard and proliferation gene expression signatures only; and d) the fourth linear predictor score is generated using genes from the 5 standard and lymph node gene expression signatures only.

95. The method of claim 87 wherein step (h) further comprises use of a microarray. Dated 16 February, 2012 Government of the United States of America, as represented by Secretary, Department of Health and Human Services Patent Attorneys for the Applicant/Nominated Person SPRUSON & FERGUSON