WO2008063521A2 - Gene-based clinical scoring system - Google Patents

Abstract

Systems and methods comprising microarray systems to assess prognostic potential in a clinical setting are described.

Description

GENE-BASED CLINICAL SCORING SYSTEM

RELATED APPLICATIONS/PATENTS & INCORPORATION BY REFERENCE

This application claims priority to U.S. provisional application Ser. No. 60/858,617, filed November 13, 2006, the entire content of which is incorporated herein by this reference.

Each of the applications and patents cited in this text, as well as each document or reference cited in each of the applications and patents (including during the prosecution of each issued patent; "application cited documents"), and each of the PCT and foreign applications or patents corresponding to and/or claiming priority from any of these applications and patents, and each of the documents cited or referenced in each of the application cited documents, are hereby expressly incorporated herein by reference.

More generally, documents or references are cited in this text, either in a Reference List before the claims, or in the text itself; and, each of these documents or references ("herein-cited references"), as well as each document or reference cited in each of the herein-cited references (including any manufacturer's specifications, instructions, etc.), is hereby expressly incorporated herein by reference.

GOVERNMENT SUPPORT The work leading to the present invention was funded in part by grant number

U54GM62119, from the National Institute of General Medical Science. Accordingly, the United States Government has certain rights to this invention.

BACKGROUND OF THE INVENTION Severe trauma is common throughout the world and leads to high morbidity and mortality. Trauma leads to long hospital stays in critical care units, and secondary complications such as the development of multiple organ failure (MOF) are common. Although the development of MOF correlates with longer ICU and hospital stays and a worse prognosis, the underlying factors that lead to its development, as well as subsequent recovery, are poorly understood.

The heterogeneity of patient outcomes following trauma complicates the development of accurate prediction of patient outcome. The current prediction of outcome after critical illness, including trauma, consists mainly of the use of scoring systems that are a composite of organ-based physiological, hematological and chemical measurements. These systems include the APACHE scores (II or III) and the Marshall and Denver organ failure scores. All use the most abnormal physiological values during a 24 hour interval in the ICU and in addition, the Marshall and Denver scores include treatment data (e.g., level of inotropes administered) in determining the score.

There are biological reasons why patients can have dramatically different outcomes after suffering a traumatic injury. Accordingly, microarray systems that measure expression of tens of thousands of genes on a chip could be useful to assess prognostic potential in a trauma setting.

SUMMARY OF THE INVENTION

The purpose and advantages of the present invention will be set forth in and become apparent from the description that follows. Additional advantages of the invention will be realized and attained by the methods and systems particularly pointed out in the written description and claims hereof, as well as from the appended figures.

To achieve these and other advantages and in accordance with the purpose of the invention, as embodied herein, the invention includes a method for assessing prognosis of a patient. The method includes measuring the expression level of a plurality of RNA species in a biological sample obtained from the patient, determining a comparative value for each RNA species by comparing the expression level for each RNA species with a reference value, and using the comparative values to compute a score indicative of the patient's condition to facilitate assessment of a prognosis for the patient.

In accordance with a further aspect of the invention, the score may be computed by performing a summation operation using the comparative values or by performing a summation operation using values derived from the comparative values. Preferably, at least one mathematical operation is performed on the comparative values prior to performing the summation operation.

In accordance with another aspect of the invention, a score may be computed at least in part by selecting a maximum comparative value from a set of comparative values. The score may be computed by performing at least one mathematical operation on the result of the summation operation. Suitable operations include, for example, taking the logarithm of the result of the summation operation, taking the square root of the result of the summation operation, and multiplying the result of the summation operation by a constant. If desired, each comparative value may be computed by performing at least one of subtracting the reference value from the expression level, squaring the expression level and dividing the expression level by the square of a standard deviation relating to the reference value. In accordance with a further aspect of the invention, a score as described herein may be computed for a population of patients, and treatment for each of the patients may be prioritized based on the score of each patient. For example, a patient with a higher score may be given treatment priority over a patient with a lower score. Preferably, the measuring of the patient's expression levels occurs within about 12 hours of an injury to the patient. hi accordance with a further aspect of the invention, the expression level for each RNA species is measured using a microarray. The microarray may include a whole genome expression array. For example, the microarray may be selected from the group consisting of: Human Genome Survey Microarray v2.0 (Applied Biosystems), Human Genome Ul 33 Plus 2.0 Array (Affymetrix), Gene Chip® Human X3P Array

(Affymetrix), Whole Human Genome Oligo Microarray Kit (Agilent Technologies), CodeLink® Human Whole Genome Bioarray (GE Healthcare), Sentrix® Human-6 Expression BeadChip (Illumina), Sentrix® HumanRef-8 Expression BeadChip (Illumina), and Human 4OK OciChip™ (Ocimum Biosolutions). If desired, the microarray may include a selected expression array selected from the group consisting of: HuGeneFL Array (Affymetrix), Human Gene Cancer Gl 10 Array (Affymetrix), Human Genome Focus Array (Affymetrix), Human Genome Ul 33 Set (Affymetrix), Human Genome U95 Set (Affymetrix), Human IA Oligo Microarray Kit (V2) (Agilent Technologies), Custom Gene Expression Microarray (Agilent Technologies), CatalogArray (CombiMatrix), CodeLink UniSet® Human 20KI (GE Healthcare), HumanTox-16 BeadChip (Illumina), HumanTox-96 Array Matrix (Illumina), and Custom OciChip arrays (Ocimum Biosolutions).

Preferably, the method further includes the step of obtaining the microarray. Moreover, the RNA species preferably correspond to nucleic acid molecules encoding a functional peptide selected from the group consisting of cytokines and antioxidants.

The method as described herein can be applied to patients with any of a myriad of conditions. For example, the patient may be experiencing trauma, shock, burns, and the like. In accordance with still a further aspect of the invention, the biological sample preferably comprises blood cells. For example, the blood cells may comprise white blood cells.

The invention also provides a kit for assessing prognosis of a patient. The kit includes a computer readable medium that comprises information corresponding to one or more RNA expression profiles of a biological sample from the patient and RNA expression profiles of a reference sample. The information also includes RNA expression profile analysis software capable of being loaded into the memory of a computer system. In accordance with a further aspect of the invention, the computer readable medium preferably includes a microarray. If desired, the kit may further comprise instructions for use.

The invention also provides a system for assessing prognosis of a patient. The system includes means for measuring the expression level of a plurality of RNA species in a biological sample obtained from the patient. The system further includes means for determining a comparative value for each RNA species by comparing the expression level for each mRNA species with a reference value. The system also includes means for computing a score indicative of the patient's condition using the comparative values. In further accordance with the invention, the system may further include means for displaying the score to a user to facilitate assessment of a prognosis for the patient. If desired, the means for computing a score computes the score at least in part by performing a summation operation using the comparative values or values derived from the comparative values. By way of further example, the means for computing a score computes the score by performing at least one mathematical operation on the comparative values prior to performing the summation operation. The mathematical operation may include, for example, multiplying the comparative value by itself at least once and multiplying the comparative value by a product of a standard deviation relating to the reference value. In accordance with a preferred embodiment, the product of the standard deviation is an inverse of a square of the standard deviation. In further accordance with the invention, the score may be displayed on a portable electronic device. In accordance with a preferred embodiment, the means for determining a comparative value is adapted and configured to access a database having a plurality of reference values stored thereon. The database may be accessed by the system over the internet.

In accordance with a further aspect of the invention, the means for determining a comparative value may computes at least one comparative value by subtracting the reference value from the expression level to create a first intermediate value, squaring the first intermediate value to create a second intermediate value, and dividing the second intermediate value by the square of a standard deviation relating to the reference value to create a third intermediate value.

In accordance with still another aspect of the invention, the system may be adapted and configured to compute a score indicative of a patient's condition for a population of patients. Accordingly, if desired, the system may further include means for prioritizing treatment for each of the patients in the population based on the score of each patient. For example, a patient with a higher score may be given treatment priority over a patient with a lower score. However, it will be recognized that a patient with a lower score may be given priority in certain situations, such as where another patient's score is so high that survival is highly unlikely, even with medical treatment. In further accordance with the invention, the means for measuring the expression level may include at least one microarray, as described herein.

The invention also provides a machine readable program, tangibly embodied on a computer readable medium, containing instructions for controlling a system for predicting clinical outcome of a patient. The instructions include means for determining a comparative value for each of a plurality of RNA species extracted from a patient to be assessed by comparing an expression level for each RNA species with an associated reference value. The instructions also include means for computing a score indicative of the patient's condition using the comparative values to facilitate assessment of a prognosis for the patient.

In accordance with a further aspect of the invention, the means for computing a score may compute the score at least in part by performing a summation operation using the comparative values or values derived from the comparative values. If desired, the means for computing a score may compute the score by performing at least one mathematical operation on the comparative values prior to performing the summation operation. The mathematical operation may be chosen from the group including multiplying the comparative value by itself at least once, and multiplying the comparative value by a product of a standard deviation relating to the reference value. The product of the standard deviation may be an inverse of a square of the standard deviation.

In further accordance with the invention, the means for determining a comparative value may be adapted and configured to access a database having a plurality of reference values stored thereon. If desired, the means for determining a comparative value may be adapted and configured to access the database through the internet. By way of further example, the means for determining a reference value may compute at least one reference value by subtracting the reference value from the expression level to create a first intermediate value, squaring the first intermediate value to create a second intermediate value, and dividing the second intermediate value by the square of a standard deviation relating to the reference value to create a third intermediate value.

In further accordance with the invention, the program may include means for computing a score indicative of a patient's condition for a population of patients. Accordingly, if desired, the program may further include means for prioritizing treatment to determine treatment priority for each of the patients in the population based at least in part on the score of each patient. The means for prioritizing treatment may also determine treatment priority based on the particular medical condition of the patient. The means for prioritizing treatment also may determine treatment priority based on the severity of the medical condition of the patient. If desired, a patient with a higher score may be assigned treatment priority in a queue over a patient with a lower score by the means for prioritizing treatment. If further desired, the program may further include means for periodically updating the queue, hi further accordance with the invention, the program may be adapted and configured to display the score on a portable electronic device. The program may also be adapted and configured to display the queue on a portable electronic device.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and are intended to provide further explanation of the invention claimed.

The accompanying figures, which are incorporated in and constitute part of this specification, are included to illustrate and provide a further understanding of the method, system and machine readable program of the invention. Together with the description, the drawings serve to explain the principles of the invention.

BRIEF DESCRIPTION OF THE FIGURES The following Detailed Description, given by way of example, but not intended to limit the invention to specific embodiments described, may be understood in conjunction with the accompanying drawings, incorporated herein by reference.

Various preferred features and embodiments of the present invention will now be described by way of non-limiting example and with reference to the accompanying drawings in which:

Figure 1 provides a conceptual illustration of the DFR score. The genomic profiles of healthy controls (pluses on the left) cluster around their mean (the dot), which becomes the reference profile. Each star on the right represents the genomic profile of a patient. The distance from the reference profile to the patient's genomic profile represents his/her DFR score (i.e., the actual computation of "distance" is not limited to

Euclidean distance).

Figure 2a graphically depicts sensitivity plotted against 1 -specificity of the predictor - the curved line denotes a very good predictor, while the straight line denotes little more than a random guess. Figure 2b graphically depicts the actual sensitivity- specificity graphs /ROC curves for DFR and APACHE II as predictors of Marshall

MOF.

Figure 3 schematically depicts a representative embodiment of a system that can be used to diagnose and triage patients in accordance with certain embodiments of the invention.

DETAILED DESCRIPTION OF THE INVENTION

L Definitions

The term "biological sample" is meant to include any sample obtained from a patient. Examples include blood, urine and tissue samples.

A "control sample" is a biological sample obtained from 1) a healthy subject ("control subject"); or 2) a subject having an underlying disease ("control subject with disease"), wherein the sample is used to establish a base level for assessing or monitoring a new complication/symptom in a test subject also having the underlying disease, the complication/symptom not being present in the control subject with disease.

The term "expression level" refers to the amount of RNA expression in a biological sample. The expression level can be either a "base" level from the microarray scan or an "adjusted" or "normalized" level determined by further data processing.

The term "expression profile" refers to a representative collection of RNA species expressed in a patient sample. Information corresponding to an expression profile can be stored by electronic means.

The term "RNA", as used herein, includes double-stranded RNA, single-stranded RNA, isolated RNA, such as partially purified RNA, essentially pure RNA, synthetic RNA, mRNA, small non-coding RNA (e.g. microRNA), recombinantly produced RNA, as well as altered RNA that differs from naturally occurring RNA by the addition, deletion, substitution and/or alteration of one or more nucleotides. Nucleotides of the RNA molecules can also comprise non-standard nucleotides, such as non-naturally occurring nucleotides, chemically synthesized nucleotides or deoxynucleotides.

The term "mRNA" refers to messenger RNA. mRNA comprises an RNA molecule transcribed from a gene, and from which a peptide is translated by the action of ribosomes. "RNA" is a molecule comprising at least one or more ribonucleotide residues. A "ribonucleotide" is a nucleotide with a hydroxyl group at the 2' position of a beta-D-ribofuranose moiety.

A "microarray" refers to a collection of nucleic acid molecules (e.g., oligonucleotides or complementary DNAs), attached to a solid support, such as a membrane, filter, chip, bead, polymer, silicon wafer or glass, and used to simultaneously analyze the expression levels corresponding genes. The nucleic acid molecules can be differentiated from each other according to their relative location (e.g., an array can include different nucleic acid molecules that are each located at a different identifiable location on a substrate).

A "nucleotide sequence" is a strand of linked nucleic acid molecules. The term "nucleic acid molecule" is well known in the art. A "nucleic acid" as used herein will generally refer to a molecule (i.e., a strand) of DNA, RNA or a derivative or analog thereof, comprising a nucleobase. A nucleobase includes, for example, a naturally occurring purine or pyrimidine base found in DNA (e.g., an adenine, guanine, thymine or cytosine) or RNA (e.g., an adenine, guanine, uracil or cytosine). The term "obtaining" as in "obtaining the microarray" is intended to include purchasing, synthesizing or otherwise acquiring the microarrays of the invention.

As used herein, "prognosis" refers to a prediction of the probable course and outcome of a disease or other medical condition and the likelihood of recovery from such a disease or other medical condition.

A "reference sample" is a combination of control samples.

A "selected expression array" is a microarray comprising a collection of nucleic acid molecules which have been subject to a selection process.

A "whole genome expression array" is a microarray comprising a collection of nucleic acid molecules representative of complete genomic expression.

In this disclosure, "comprises," "comprising," "containing" and "having" and the like can have the meaning ascribed to them in U.S. Patent law and can mean " includes," "including," and the like; "consisting essentially of or "consists essentially" likewise has the meaning ascribed in U.S. Patent law and the term is open-ended, allowing for the presence of more than that which is recited so long as basic or novel characteristics of that which is recited is not changed by the presence of more than that which is recited, but excludes prior art embodiments. IL Systems, Methods and Machine Readable Programs

Reference will now be made in detail to the present preferred embodiments of the invention, examples of which is illustrated in the accompanying drawings. The method and corresponding steps of the software program of the invention will be described in conjunction with the detailed description of the system below.

The systems, methods and machine readable programs embodied herein may be used for facilitating the prognosis of a patient. The present invention is particularly suited for facilitating triage of a plurality of individuals that require medical treatment.

In accordance with the invention, a system for assessing prognosis of a patient is provided. The system includes means for measuring the expression level of a plurality of RNA species in a biological sample obtained from the patient, and means for determining a comparative value for each RNA species by comparing the expression level for each mRNA species with a reference value. The system further includes means for computing a score indicative of the patient's condition using the comparative values, and means for displaying the score to a user to facilitate assessment of a prognosis for the patient. For purpose of explanation and illustration, and not limitation, a view of an exemplary embodiment of the system in accordance with the invention is shown in Fig. 3 and is designated generally by reference character 100.

As depicted in Fig. 3, system 100 includes a means for measuring the expression level of a plurality of RNA species in a biological sample obtained from a patient, for example, in the form of a microarray 110. Suitable microarrays 110 are discussed in detail in Section III below.

As further depicted in Fig. 3, a computer network 120 is further provided below that is adapted and configured to read and process data from microarray 110. Those skilled in the art will readily appreciate that a system 100 in accordance with the present disclosure may include the various computer and network related software and hardware typically used in a distributed computing network, that is, programs, operating systems, memory storage devices, input/output devices, data processors, servers with links to data communication systems, wireless or otherwise, such as those which take the form of a local or wide area network, and a plurality of data transceiving terminals within the network, such as personal computers. Those skilled in the art will further appreciate that, so long as its users are provided local and remote access to a system in accordance with the present disclosure, the precise type of network and associated hardware are not vital to its full implementation. Preferably, graphical user interfaces (GUIs) used by the present system incorporate user- friendly features and fit seamlessly with other operating system interfaces, that is, in a framed form having borders, multiple folders, toolbars with pulldown menus, embedded links to other screens and various other selectable features associated with animated graphical representations of depressible buttons. These features can be selected (i.e., "clicked on") by the user via connected mouse, keyboard, voice command or other commonly used tool for indicating a preference in a computerized graphical interface.

Accordingly, as further embodied herein, computer network 120 may include a plurality of stationary and/or mobile computers 122 such as laptop computers and/or personal digital assistants having associated terminals 124 and graphical user interfaces 126 as known in the art connected to one or more server computers 115 by appropriate hardware. At least one computer 122 in system is equipped with suitable software determining a comparative value for each RNA species by comparing the expression level for each mRNA species with a reference value.

As is further depicted in Fig. 3, a database 128 is provided for storing reference values. Any suitable computer 122 as described herein can access database in any known manner, such as through local networks and/or over the internet, to access reference values to permit comparison with the expression levels obtained for each mRNA species. Storing reference values in a central location such as database can be advantageous because it permits ready supplementation and updating of the reference values with additional data. As the amount of data expands in database 128 (such as by increasing the population of individuals upon which the reference data is based, or by adding additional reference points to new nucleic acid molecules) additional permutations of reference points will become possible to diagnose a given patient. It is also possible to provide a plurality of databases, including a database 128 at a location proximate a treatment location (such as within the hospital) that may be updated periodically by reference to a larger central database over the internet.

Any suitable mathematical method can be used to compute a comparative value to help determine if the expression level for a given mRNA sequence is meaningfully different from a reference value. For example, a comparative value may be computed by simply subtracting the reference value from the expression level. This simple comparative value can be used to compute a score to facilitate diagnosing the patient, or can alternatively be used as a first intermediate value that can be further manipulated mathematically to arrive at a comparative value. For example, the first intermediate value can be multiplied by itself any suitable number of times (squaring, cubing, etc.) to arrive at a comparative value that can be used to directly compute a clinical score. Alternatively, the first intermediate value may be multiplied by itself to create a second intermediate value that can be operated on further. If a second intermediate value is created, the second intermediate value can be multiplied or divided by a statistically derived value such as a weighting factor. The weighting factor can be, for example, the inverse of the square of a standard deviation relating to the reference value to arrive at the comparative value. This value can similarly be used to directly compute a clinical score, or can be treated as yet a third intermediate value further operated upon mathematically to arrive at a score. It will be appreciated that any other suitable mathematical operations may be employed to arrive at a suitable comparative value, in accordance with the teachings herein. For example, the mathematical operations described above can be performed on a simple comparative value (such as one obtained only by subtraction) alone, or in combination as described above.

The system further includes means for computing a score indicative of the patient's condition using the comparative values, and means for displaying the score to a user to facilitate assessment of a prognosis for the patient.

For purposes of illustration and not limitation, as embodied herein and as depicted in Fig. 3, computer 122 preferably accesses the reference values from database 128 or other suitable location (such as from within computer 122), and performs suitable mathematical operations thereon to arrive at one or more comparative values, as described above. Computer 122 preferably runs suitable software for performing further mathematical operations on the derived comparative values to compute a score for a patient that can be used to facilitate assessment of the patient's prognosis.

Computer 122 may compute a score in a variety of manners. For example, the score may be computed at least in part by performing a summation operation using the comparative values that have previously been established according to any suitable method. The summation operation may simply add the previously identified comparative values, or may include additional or alternative steps. The summation operation may include or be followed by additional mathematical operations. Further operations may be applied to the result of the summation, such as simply multiplying the result of the summation by a constant or by applying a mathematical function (e.g., the square root or the log base 10 or the natural logarithm) to the result of the summation to arrive at a score.

It is also possible for computer 122 to compare the comparative values to one another in order to arrive at a score. For example, the maximum comparative value can be used to compute a score by using the maximum comparative value itself as a score, or by performing a further mathematical operation on it, such as multiplying the maximum comparative value by a constant, or by applying a mathematical function (e.g., the square root or the log base 10 or the natural logarithm) to the maximum value to arrive at a score. Alternatively, a subset of the comparative values may be used to compute a score. A subset of the set of comparative values may be identified by selecting a predetermined number (e.g., 2, 3, 4, 5, and so on) of the comparative values that are the maximum comparative values within the set. These values may be further manipulated, such as by performing a summation operation as described above on the values.

By way of further example, the score may be arrived at in a number of other manners. For example, for each probeset, the patient's expression level can be multiplied by the reference value to arrive at a first comparative value for each probeset. All first comparative values can then be summed to arrive at a first sum A. Next, each patient's expression level for each probeset can be squared to arrive at a second comparative value for each probeset. These second comparative values may then be summed to arrive at a second sum B. Next, each reference probeset can be squared to arrive at a comparative reference value for each reference probeset. These values can also be summed to arrive at a third sum C. Next, first sum A may be divided by the square root of the product of second sum B with third sum C. Such a score provides an angular difference between patient and reference profiles. By way of still further example, the immediately preceding method can be performed on one or more subsets of the probesets. For example, the method in the preceding paragraph can be performed on a first probe subset Y (e.g., upregulated patient probesets or just the cytokines). This method can also be carried out on a second probe subset Z (e.g., downregulated probesets or only antioxidants). Both resulting values may be reported as the score. Alternatively, the difference may be taken between the subset Y score and subset Z score to arrive at a final score. By way of further example, the maximum of the subset Y and subset Z scores may be reported as the score. It will be appreciated that this embodiment of the method, system and machine readable program can be implemented on any suitable number of subsets (e.g., 3, 4, 5, 6, 7, and so on).

Comparative values may be arrived at in a number of other manners. For example, the patient status may be characterized by computing the angle between the vector formed by a reference profile and the vector formed by a patient's expression profile. Furthermore, if desired, both the reference profile and patient's expression profile may be treated as functions and any functional analytic or signal processing measure of function difference can be applied to produce a score.

Once a score is obtained for a patient, system 100 preferably displays the score for assessment by medical personnel to facilitate determining prognosis of the patient. While the score may be displayed on any suitable computer 122 in system 100, in accordance with one embodiment the score is displayed on a portable electronic device, such as a personal digital assistant adapted and configured to run software to assist in diagnosing patients. Displaying such data on a portable device that may be carried around, for example, by emergency room personnel, can permit easy access to data as new patients are processed.

Preferably, system 100 is adapted and configured to compute a score indicative of a patient's condition for a population of patients, such as a population of patients in a triage environment, such as an emergency room or even a battlefield condition. If so equipped, system 100 is also preferably provided with means for prioritizing treatment for each of the patients in the population based on the score of each patient. This can be implemented, for example, by a software program that is adapted and configured to rank patients in a queue according to their score, and other suitable parameters, if appropriate. Preferably, a patient with a higher score is given treatment priority over a patient with a lower score. In other embodiments, a score may be so high as to indicate a low prognosis for survival, in which case, a patient with a lower score is given treatment priority over a patient with a the high score. Accordingly, As embodied herein, the relative magnitude of a patient's score generally describes the extent to which a patient differs from a statistically significant population on a genetic level. As such, a patient with a higher score may be genetically predisposed to weaker health and have a weaker constitution. In such a situation, the prognosis of a patient with a medical condition that may not be perceived to be as serious as that as another patient may be changed significantly by considering the genetic level information.

In accordance with another embodiment, the system 100 and associated machine readable program may prioritize treatment based upon the computed score in combination with the particular medical condition of the patient, including the severity of the patient's medical condition.

Advantageously, if system 100 is implemented over a local area network in a hospital, a plurality of medical professionals can be provided with information about all patients in the emergency room, or even in the hospital as a whole or other hospitals. The treatment priority of all patients within a population of interest can be organized into a queue that may be updated periodically (e.g., every minute, every five minutes, etc.) and displayed on a portable device in the possession of each professional to permit all professionals in the facility to devote limited emergency room resources to patients that need treatment the most. Similarly, nearby hospitals can be networked together in system 100 to help medical professionals decide whether patients from a first overloaded emergency room should be re-routed to a hospital that is comparatively underutilized. Block diagrams and other representations of circuitry of the system and machine readable program embodied herein represent conceptual views of illustrative circuitry and software embodying the principles of the invention. Thus the functions of the various elements shown in the Figures and as otherwise described herein may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. The functions of those various elements may be implemented by, for example, digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read-only memory (ROM) for storing software, random access memory (RAM), and non-volatile storage. Other hardware, conventional and/or custom, may also be included.

In the claims hereof any element expressed as a means for performing a specified function is intended to encompass any way of performing that function including, for example, a) a combination of circuit elements which performs that function or b) software in any form, including, therefore, firmware, microcode or the like, combined with appropriate circuitry for executing that software to perform the function. The invention as defined by such claims resides in the fact that the functionalities provided by the various recited means are combined and brought together in the manner which the claims call for. Applicants thus regard any means which can provide those functionalities as equivalent to those shown herein.

Similarly, it will be appreciated that the system flows described herein represent various processes which may be substantially represented in computer-readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown. Moreover, the various processes can be understood as representing not only processing and/or other functions but, alternatively, as blocks of program code that carry out such processing or functions. III. Microarravs

A DNA microarray (also known as gene or genome chip, DNA chip, or gene array) is a collection of microscopic DNA spots attached to a solid surface, such as glass, plastic, or silicon chip forming an array for the purpose of expression profiling, monitoring expression levels for thousands of genes simultaneously. The affixed DNA segments are known as probes, thousands of which can be present in a single DNA microarray.

Microarray Fabrication

Microarrays of the invention can be fabricated using a variety of techniques known in the art, including printing with fine-pointed pins onto glass slides, photolithography using pre-made masks, photolithography using dynamic micromirror devices, ink-jet printing (Lausted, et al. 2004 Genome Biol 5:R58), or electrochemistry on microelectrode arrays.

For example, microarray fabrication described by Affymetrix typically combines semiconductor fabrication techniques, solid phase chemistry, combinatorial chemistry, molecular biology, and sophisticated robotics to yield a photolithographic manufacturing process that produces DNA microarrays with millions of probes on a small glass chip.

The photolithographic process begins by coating a 5" x 5" quartz wafer with a light-

sensitive chemical compound that prevents coupling between the wafer and the first nucleotide of the DNA probe being created. Lithographic masks are used to either block or transmit light onto specific locations of the wafer surface. The surface is then flooded with a solution containing either adenine, thymine, cytosine, or guanine, and coupling occurs only in those regions on the glass that have been deprotected through illumination. The coupled nucleotide also bears a light-sensitive protecting group, so the cycle can be repeated. In this way, the microarray is built as the probes are synthesized through repeated cycles of deprotection and coupling. The process is repeated until the probes reach their full length, usually 25 nucleotides.

As another example, microarrays described by GE Healthcare are based on a 3-D aqueous gel matrix slide surface with 30-base oligonucleotide probes. Probe attachment is accomplished through covalent interaction between the amine-modified 5' end of the dispensed oligonucleotide and the activated functional group present in the gel matrix. The 3-D gel matrix provides an aqueous environment that holds the probe away from the surface of the slide, allowing for maximum interaction between probe and target.

In microarray fabrication, an array of oligonucleotides may be synthesized on a solid support. Exemplary solid supports include glass, plastics, polymers, metals, metalloids, ceramics, organics, etc. Using chip masking technologies and photoprotective chemistry, it is possible to generate ordered arrays of nucleic acid probes. Nucleic acid probes may be obtained, for example, by PCR amplification of gene segments from genomic, cDNA (e.g., RT-PCR), or cloned sequences. cDNA probes may be prepared according to methods known in the art and further described herein, for example, by reverse-transcription PCR (RT-PCR) of RNA using sequence specific primers. Sequences of genes or cDNA from which probes are generated may be obtained, for example, from GenBank, other public databases, or publications. Oligonucleotide probes may also be synthesized by standard methods known in the art, for example, by automated DNA synthesizer or any other chemical method. Nucleic acid probes may be natural nucleic acids or chemically modified nucleic acids (e.g., composed of nucleotide analogs); however, the probes should possess activated hydroxyl groups compatible with the linking chemistry. The protective groups may be photolabile, or the protective groups may be labile under certain chemical conditions (e.g., acid). The surface of the solid support may contain a composition that generates acids upon exposure to light. Thus, exposure of a region of the substrate to light generates acids in that region that remove the protective groups in the exposed region. Also, the synthesis method may use 3'-protected 5'-0-phosphoramidite-activated deoxynucleoside. In this case, the oligonucleotide is synthesized in the 5' to 3' direction, which results in a free 5' end. Arrays may also include control and reference nucleic acids. Control nucleic acids include, for example, prokaryotic genes such as bioB, bioC and bioD, ere from Pl bacteriophage or polyA controls, such as dap, lys, phe, thr, and trp. Reference nucleic acids allow the normalization of results from one sample analysis to another and the comparison of multiple sample analyses on a quantitative level. Exemplary reference nucleic acids include housekeeping genes of known expression levels, for example, GAPDH, hexokinase, and actin. Types of Microarrays

DNA microarrays can be used to detect RNAs that may or may not be translated into active proteins, i.e., for expression analysis/profiling. Since there can be tens of thousands of distinct reporters on an array, each microarray experiment can accomplish the equivalent number of genetic tests in parallel.

In spotted microarrays (or two-channel microarrays), the probes are oligonucleotides, cDNA, or small fragments of PCR products corresponding to mRNAs. The spotted microarray is typically hybridized with cDNA from two samples to be compared (e.g., patient and control) that are labeled with two different fluorofores. The samples can be mixed and hybridized to one single microarray that is then scanned, allowing the visualization of up-regulated and down-regulated genes in one go.

In oligonucleotide microarrays (or single-channel microarrays), the probes are designed to match parts of the sequence of known or predicted mRNAs. There are commercially available designs that cover the complete human genome, including, without limitation: Human Genome Survey Microarray v2.0 (Applied Biosystems), Human Genome Ul 33 Plus 2.0 Array (Affymetrix), Gene Chip® Human X3P Array (Affymetrix), Whole Human Genome Oligo Microarray Kit (Agilent Technologies), CodeLink® Human Whole Genome Bioarray (GE Healthcare), Sentrix® Human-6 Expression BeadChip (Illumina), Sentrix® HumanRef-8 Expression BeadChip (Illumina), and Human 4OK OciChip™ (Ocimum Biosolutions). These microarrays give estimations of the absolute value of gene expression. Thus, "whole genome" arrays provide a one-array view comprehensive (as comprehensive as the state of the research art allows) of whole human genome expression.

There are also commercially available designs that cover subsets of human genes, including, without limitation: HuGeneFL Array (Affymetrix), Human Gene Cancer Gl 10 Array (Affymetrix), Human Genome Focus Array (Affymetrix), Human Genome Ul 33 Set (Affymetrix), Human Genome U95 Set (Affymetrix), Human IA Oligo Microarray Kit (V2) (Agilent Technologies), Custom Gene Expression Microarray (Agilent Technologies), CatalogArray (CombiMatrix), CodeLink UniSet® Human 20KI (GE Healthcare), HumanTox- 16 BeadChip (Illumina), HumanTox-96 Array Matrix (Illumina), and Custom OciChip arrays (Ocimum Biosolutions). "Subset genome" or "selected" expression arrays allows monitoring of a specific subset of human genes. In one embodiment, the selected genes of the microarray are cytokines, including but not limited to PAF, N-formylated peptides, C5a, LTB4 and LXA4, chemokines: CXC, IL-8, GCP-2, GRO, GROα, GROβ, GROγ, ENA-78, NAP-2, IP-IO, MIG, I-TAC, SDF-Io, BCA-I, PF4, Bolekine, MIP- lα, MIP-I β, RANTES, HCC-I, MCP-I, MCP-2, MCP-3, MCP-4, MCP-5 (mouse), Leukotactin-1 (HCC-2, MIP-5), Eotaxin, Eotaxin-2 (MPIF2), Eotaxin-3 (TSC), MDC, TARC, SLC (Exodus-2, όCKine), MIP-3α (LARC, Exodus-1), ELC (MJP-3β), 1-309, DC-CKl (PARC, AMAC-I), TECK, CTAK, MPIFl (MIP-3), MIP-5 (HCC-2), HCC-4 (NCC-4), C Lymphotactin, and CX3C Fracktelkine (Neurotactin) concentration gradients. The cytokine can be a member of the Cys-X-Cys family of chemokines (e.g., chemokines that bind to the CXCR-4 receptor). Preferred cytokines of the invention include SDF-lα, SDF-lβ, met-SDF-lβ, IL-I, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-IO, IL-12, IL-15, IL-18, TNF, IFN-α, IFN-β, IFN-γ, granulocyte- macrophage colony stimulating factor (GM-CSF), granulocyte colony stimulating factor (G-CSF), macrophage colony stimulating factor (M-CSF), TGF-β, FLT-3 ligand, VEGF, DMDA, endothelin, and CD40 ligand.

In another embodiment, the selected genes of the microarray are antoxidants, including but not limited to genes expressed in response to oxidative stress (such as those of Human Genome Ul 33 Set (Affymetrix).

Oligonucleotide arrays can, for example, be produced by piezoelectric deposition with full-length oligonucleotides or by in situ synthesis. Long Oligonucleotide Arrays are composed of 60-mers, or 50-mers and can, for example, be produced by ink-jet printing on a silica substrate. Short Oligonucleotide Arrays are composed of 25-mer or 30-mer and can, for example, be produced by photolithographic synthesis on a silica substrate or piezoelectric deposition on an acrylamide matrix. Genomic Analysis via Microarrays

Generally, determining expression profiles with microarrays involves obtaining an mRNA sample from a subject and preparing labeled nucleic acids therefrom, contacting the target nucleic acids with an array under conditions sufficient for the target nucleic acids to bind to the corresponding probes on the array, for example, by hybridization or specific binding, optional removal of unbound targets from the array, detecting the bound targets, and analyzing the results, for example, using computer- based analysis methods. Nucleic acid specimens may be obtained from an individual to be tested using either "invasive" or "non-invasive" sampling means. A sampling means is said to be "invasive" if it involves the collection of nucleic acids from within the skin or organs of a subject. Examples of invasive methods include blood collection, semen collection, needle biopsy, pleural aspiration, umbilical cord biopsy, etc. Examples of such methods are discussed by Kim, et al, (J. Virol. 66:3879-3882, 1992); Biswas, et al, {Ann. NY Acad. Sci. 590:582-583, 1990); and Biswas, et al, (J. Clin. Microbiol. 29:2228-2233, 1991). A specific subset of cells (for example, leukocytes) may be further isolated from the sample (for example, via buffy coat isolation or lysis isolation). In embodiments of the present invention, one or more cells from the subject to be tested are obtained, and RNA is isolated from the cells. RNA may be extracted from tissue or cell samples by a variety of methods, for example, guanidium thiocyanate lysis followed by CsCl centrifugation (Chirgwin, et al., Biochemistry 18:5294-5299, 1979). RNA may be amplified by methods known in the art (e.g., RT-PCR, cRNA synthesis). Generally, the target molecules will be labeled to permit detection of hybridization to a microarray. Examples of directly detectable labels include isotopic and fluorescent moieties incorporated, usually by a covalent bond, into a moiety of the probe, such as a nucleotide monomelic unit (e.g., dNMP of the primer), or a photoactive or chemically active derivative of a detectable label that can be bound to a functional moiety of the probe molecule. Fluorescent moieties or labels of interest include coumarin and its derivatives (e.g., 7-amino-4-methylcoumarin, aminocoumarin); bodipy dyes such as Bodipy FL and cascade blue; fluorescein and its derivatives (e.g., fluorescein isothiocyanate, Oregon green); rhodamine dyes (e.g., Texas red, tetramethylrhodamine); eosins and erythrosins; cyanine dyes (e.g., Cy2, Cy3, Cy3.5, Cy5, Cy5.5, Cy7); FluorX, macrocyclic chelates of lanthanide ions (e.g., quantum dye.TM.); fluorescent energy transfer dyes such as thiazole orange-thidium heterodiimer, TOTAB, dansyl, etc. Chemiluminescent labels include luciferin and 2,3- dihydrophthalazinediones, for example, luminol.

Labels may also be members of a signal producing system that act in concert with one or more additional members of the same system to provide a detectable signal. Illustrative of such labels are members of a specific binding pair, such as ligands, for example, biotin, fluorescein, digoxigenin, antigen, polyvalent cations, chelator groups and the like. Members may specifically bind to additional members of the signal producing system, and the additional members may provide a detectable signal either directly or indirectly, for example, an antibody conjugated to a fluorescent moiety or an enzymatic moiety capable of converting a substrate to a chromogenic product (e.g., alkaline phosphatase conjugate antibody and the like). In a preferred embodiment, biotinylated cRNA is generated from the extracted RNA.

To compare expression levels, labeled nucleic acids may be contacted with the array under conditions sufficient for binding between the target nucleic acid and the probe on the array. Ln one embodiment, the hybridization conditions may be selected to provide for the desired level of hybridization specificity; that is, conditions sufficient for hybridization to occur between the labeled nucleic acids and probes on the microarray.

Hybridization may be carried out in conditions permitting essentially specific hybridization. The length and GC content of the nucleic acid will determine the thermal melting point and thus, the hybridization conditions necessary for obtaining hybridization. These factors are well known to a person of skill in the art, and may also be tested in assays. An extensive guide to nucleic acid hybridization may be found in Tijssen, et al. (Laboratory Techniques in Biochemistry and Molecular Biology, Vol. 24: Hybridization With Nucleic Acid Probes, P. Tijssen, ed. Elsevier, N. Y., (1993)).

Non-specific binding or background signal may be reduced by the use of a detergent (e.g, C-TAB) or a blocking reagent (e.g., sperm DNA, cot-1 DNA, etc.) during the hybridization.

The production of hybridization patterns of labeled target nucleic acids on the array surface results. The hybridization patterns of labeled nucleic acids may be visualized or detected in a variety of ways, with the particular manner of detection selected based on the particular label of the target nucleic acid. Representative detection means include scintillation counting, autoradiography, fluorescence measurement, colorimetric measurement, light emission measurement, light scattering, and the like.

One such method of detection utilizes an array scanner (Affymetrix, Santa Clara, Calif). This scanner is controlled from a system computer with an interface and easy- to-use software tools. The output may be directly imported into or directly read by a variety of software applications. Scanning devices are described in, for example, U.S. Pat. Nos. 5,143,854 and 5,424,186.

Following the data gathering operation, the data will typically be reported to a data analysis operation. To facilitate the sample analysis operation, the data obtained by the reader from the device may be analyzed using a digital computer. Typically, the computer will be appropriately programmed for receipt and storage of the data from the device, as well as for analysis and reporting of the data gathered, for example, subtraction of the background, deconvolution of multi-color images, flagging or removing artifacts, verifying that controls have performed properly, normalizing the signals, interpreting fluorescence data to determine the amount of hybridized target, normalization of background and single base mismatch hybridizations, and the like.

A system may comprise a search function that allows one to search for specific patterns, for example, patterns relating to differential gene expression, for example, between the expression profile of a cell from a trauma patient and a reference expression profile. Various algorithms are available for analyzing gene expression profile data, such as described herein above.

The present invention is additionally described by way of the following illustrative, non-limiting Examples that provide a better understanding of the present invention and of its many advantages.

EXAMPLES Example 1

To determine whether gene expression in circulating leukocyte RNA can be utilized very early in resuscitation after trauma injury to predict outcome, genomic data derived from blood samples from 80 patients with traumatic injury and 26 healthy volunteers was used to calculate a score for each patient. The score was designed to capture the patient's genomic deviation from a reference profile (DFR score). The predictive value of the DFR score was assessed using various outcome measures, including multiple organ failure, lengths of stay in the ICU and hospital, days on the ventilator, and mortality.

Subject Recruitment

80 patients were recruited at multiple participating institutions. These patients had suffered severe blunt trauma (APACHE II scores 10-41, mean 29.7, and Injury Severity Scores 8-75, mean 26.4), but they did not have severe brain injury, defined as Glasgow Coma Score less than 9 with abnormal CT scan of the head, and they were expected to survive beyond 24 hours. Patient ages ranged from 16-54, and they were 60% male. Twenty-six healthy control subjects were recruited, ranging from 18 to 45 years in age, and 54% were male. The control subjects included 5 young adults ages 18- 20 who had been burned 3 to 6 years previously but had no other health problems.

The study was approved by the institutional review board of each participating center and written informed consent was obtained from all patients or their legal next of kin and from all control volunteers. Data Collection

For the trauma patients, peripheral blood samples were taken within 12 hours of injury and then at fixed intervals until the patient left the ICU. Total blood leukocytes for patients and controls were processed [I]. The resulting cRNA was hybridized onto Affymetrix® HU133A+ 2.0 microarrays. The microarray data was normalized and modeled using dChip software [2] to provide gene expression data.

Modified Marshall and Denver scores indicated patient clinical state during the stay in ICU. The modified Marshall score used herein omits the neurologic component and assigned 5 component scores based on the pressure-adjusted heart rate, PaO₂ /FiO₂ , creatinine, bilirubin, and platelet levels [3]. The Denver score is based on 4 component scores determined by the patient's PaO₂ /FiO₂ , creatinine, and bilirubin levels and by the level of inotropes being administered by the physician [4] and [6]. APACHE II [7] and ISS [5] were used as overall clinical measures of injury severity, in addition to numerous data items capturing details of the patient's physiological state. Clinical outcomes that occurred while the patient was in the ICU (up to 28 days post-injury) were recorded. Primary outcomes of interest were mortality, development of multiple organ failure/dysfunction (MOF), number of infections and other complications, and measures of healing rates such as length of ICU stay and hospital stay. Multiple organ failure was defined in two ways: attaining a modified Marshall MOF score of 6 or higher (Marshall MOF) and attaining a Denver score of 4 or higher (Denver MOF). Table 1, below, presents the baseline, injury and outcome data for the patients.

Data Analysis

The gene expression data for each subject was based on the initial blood sample and consisted of the set of 54675 numbers representing the expression levels of the 54675 probesets on the Affymetrix® HU133A+ 2.0 chips, normalized and modeled using the Li-Wong algorithm [2] and coefficients generated from more than 450 patient microarrays processed in the study." Each of the approximately 22000 human genes is represented by one to several probesets. The set of expressions for the 54600 probesets constituted the "genomic profile" for a subject. For the control group, the probeset-by-probeset mean and variance were computed. The set of these means formed the "healthy reference profile", essentially, the centroid of the control genomic profiles.

Construction of DFR Score

For each patient, a Deviation From Reference Score (DFR) was created by squaring the difference between the patient expression and the control group mean expression for each probeset, scaling this by the control group variance, and summing over all 14400 probesets. The result is effectively the distance between the patient's genomic profile and the healthy reference profile. Equation (1), below, shows the actual computation used to create the DFR score.

wherein e\ is the patient expression level, and Mj and Sj² are the control group mean and variance for the i^th probeset. In Equation (1), above, division by the variance in the controls is a re-scaling that prevents the DFR score from being dominated by genes that are inherently more variable or highly expressed. The natural logarithm is applied to make the distribution of the resulting DFR more symmetric over the patient population. Since the DFR score provided a measure of how different a patient's genomic profile was from the healthy reference profile, lower scores should indicate better health status. Correlation of DFR Scores with Baseline and Injury Data Table 2, below, shows all baseline and injury variables that were significantly associated either with the DFR score or with the APACHE II score.

The DFR score was not significantly associated with worst base deficit 0-12 hours post injury or with ventilation status on admission. However, this gene-based score was more highly correlated with ISS, maximum AIS score and age than APACHE, and it discriminated between patients with and without pre-existing comorbidities, whereas APACHE did not.

Association of Patient Scores with Outcome 0 Linear and logistic regression were used to test the association between various outcomes of interest and each of the 3 patient scores ~ DFR, APACHE II, ISS. In order to directly compare the strength of association of these scores with outcome, each of them was converted to a standardized score. Each type of score was transformed to a corresponding z-score as follows: 5 z-score = (score - mean score)/standard deviation of scores. Thus, for the continuous outcomes, the linear model (Equation (2) was fitted using a standardized score.

(2) outcome = α • score + ε

The regression coefficient α is proportional to the correlation between the outcome and the score, while the constant of proportionality depends only on outcome. This allows direct comparison of the models for a given outcome.

Table 3 A, below, compares linear regression models for the standardized DFR, APACHE and ISS scores. In comparing the strength of association, the patient score with higher coefficient α is more highly correlated with outcome.

Table 3A

Association of Continuous Outcomes with Patient Scores Standardized)

According to Table 3 A, above, the DFR score was more strongly associated with all of the outcomes than either of the clinical scores.

For dichotomous outcomes such as mortality and MOF, logistic regression models are employed. In brief, the log odds of the event are modeled for a given predictor as a linear function of the predictor. If p(x) is the probability of outcome occurring when the value of the predictor is x, then the log odds of outcome for predictor x = In [ p(x) / (l-p(x) ) ]. This is denoted logit (outcome | x). The logistic model is (Equation (3), below):

(3) logit (outcome | score) = α • score + ε Because the scores have been standardized, strength of association can be compared by comparing the size of α. However, logistic regression models can also be compared by comparing the odds ratios. An odds ratio shows the factor by which the odds of event increase (or decrease) as the predictor increases by one unit. An odds ratio of 1.5 signifies that the odds of the event are 1.5 times greater when the predictor increases by 1 unit.

Table 3B, below, compares logistic regression models for mortality and MOF, using the standardized DFR, APACHE and ISS scores as predictors.

Table 3B

Association of Event Outcomes with Patient Scores

The DFR score shows a stronger association with all of the outcomes than either of the clinical scores. Not only were the odds ratios higher, the models show that DFR score was a statistically significant class predictor, even when the other scores had no significant predictive power for outcome.

Table 3B, above, presents another measure of predictive power of the patient scores — the c-statistic. This statistic is indirectly related to the sensitivity and specificity of a predictor. If the sensitivity of the predictor is plotted against (1- specificity), one point for each level of predictor, a very good predictor would produce a graph similar to the curved line in Figure 2a. For this curve, points nearest the upper left corner have both high sensitivity and high specificity. In fact, the curve for a perfect predictor would climb up the y-axis to 1 and then continue as a horizontal line at y=l, with an area of 1 between it and the x axis. In contrast, when a predictor does no better than randomly guessing, a graph similar to the straight line y=x is obtained, with an area of .5 between it and the x-axis. The c-statistic shows the area under the graph of specificity vs. 1 -sensitivity. Thus, the bigger the c statistic is, the more favorable the sensitivity and specificity of the predictor are. Table 3B, above, shows that the c- statistics for the DFR score are uniformly better than for either of the two standard clinical scores. To ascertain whether the gene-based DFR score provides the same patient information contained in the APACHE II score, the association of DFR score was modeled with outcome after adjusting for the effect of APACHE score on outcome. For the continuous outcomes, the multilinear regression model (Equation (4)) was used.

(4) outcome = α • APACHE + β • DFR + ε

For the event outcomes, multiple logistic regression (Equation (5)) was used.

(5) logit (event | predictors) = α • APACHE + β • DFR + ε

Tables 4a and 4b, below, show the results for these models.

Table 4A

Association of Continuous Outcome with DFR Score after Adjustment for APACHE

Score

With regard to the continuous outcomes of Table 4a, above, for any fixed level of APACHE score, adding the DFR information significantly improved the correlation with outcome. Since standardized scores were used, the relative magnitude of the coefficients indicates the relative importance of each of APACHE and DFR in estimating the value of the outcome. The model further indicates that the additional information in the DFR score is independent of APACHE score.

To determine the contribution of including DFR to APACHE II in predicting mortality and MOF, the model significance and c statistics for the uni-predictor APACHE in Table 3b, above, can be compared with the multivariate predictor α APACHE + β DFR in Table 4b, above. In all cases, adding DFR to APACHE improves the model significance and c statistic considerably, even when the resulting multivariate model is only borderline significant. For instance, for mortality, the model significance is improved from p = .58 to p = .08, and the c statistic increases from .58 to .73. Similar predictive changes occur for the MOF events.

All statements herein reciting principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure. Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity and understanding, it will be apparent to those skilled in the art that certain changes and modifications can be practiced. Therefore, the description and examples should not be construed as limiting the scope of the invention, which is delineated by the appended numbered claims. REFERENCES

[1] Cobb JP, Mindrinos MN, Miller-Graziano C, Calvano SE, Baker HV, Xiao W et al.

Application of genome-wide expression analysis to human health and disease. Proc Natl Acad Sci U S A 2005; 102(13):4801-4806.

[2] Li C, Wong WH. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci U S A 2001; 98(l):31-36.

[3] Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23: 1638- 1652.

[4] Sauaia A, Moore FA, Moore EE, Haenel JB, Read RA, Lezotte DC. Early predictors of postinjury multiple organ failure. Arch Surg 1994; 129:39-45.

[5] Baker SP, O'Neill B, Haddon W, Jr., Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974; 14(3):187-196.

[6] Sauaia A, Moore FA, Moore EE, Lezotte DC. Early risk factors for postinjury multiple organ failure. World J Surg 1996; 20:392-400.

[7] Knaus W, Draper EA, Wagner D, Zimmerman JE. APACHE II: A severity of disease classification system. Critical Care Medicine 1985; 13:818-829.

Claims

WHAT IS CLAIMED IS

1. A method for assessing prognosis of a patient, comprising: a) measuring the expression level of a plurality of RNA species in a biological sample obtained from the patient; b) determining a comparative value for each RNA species by comparing the expression level for each RNA species with a reference value; and c) using the comparative values to compute a score indicative of the patient's condition to facilitate assessment of a prognosis for the patient.

2. The method of claim 1, wherein the score is computed by performing a summation operation using the comparative values or by performing a summation operation using values derived from the comparative values.

3. The method of claim 2, wherein at least one mathematical operation is performed on the comparative values prior to performing the summation operation.

4. The method of claim 1, wherein a score is computed at least in part by selecting a maximum comparative value from a set of comparative values.

5. The method of claim 2, wherein the score is computed by performing at least one mathematical operation on the result of the summation operation chosen from the group consisting of: a) taking the logarithm of the result of the summation operation; b) taking the square root of the result of the summation operation; and c) multiplying the result of the summation operation by a constant.

6. The method of claim 1, wherein each comparative value is computed by performing at least one of the following mathematic operations: a) subtracting the reference value from the expression level; b) squaring the expression level; and c) dividing the expression level by the square of a standard deviation relating to the reference value.

7. A method comprising: a) computing a score according to the method of claim 1 for a population of patients; and b) prioritizing treatment for each of the patients based on the score of each patient.

8. The method of claim 7, wherein a patient with a higher score is given treatment priority over a patient with a lower score.

9. The method of claim 1, wherein the measuring occurs within about 12 hours of an injury to the patient.

10. The method of claim 1, wherein the expression level for each RNA species is measured using a microarray.

11. The method of claim 10, wherein the microarray comprises a whole genome expression array.

12. The method of claim 11, wherein the microarray is selected from the group consisting of: Human Genome Survey Microarray v2.0 (Applied Biosystems), Human Genome Ul 33 Plus 2.0 Array (Affymetrix), Gene Chip® Human X3P Array (Affymetrix), Whole Human Genome Oligo Microarray Kit (Agilent Technologies), CodeLink® Human Whole Genome Bioarray (GE Healthcare), Sentrix® Human-6 Expression BeadChip (Illumina), Sentrix® HumanRef-8 Expression BeadChip (Illumina), and Human 4OK OciChip™ (Ocimum Biosolutions).

13. The method of claim 10, wherein the microarray comprises a selected expression array.

14. The method of claim 13, wherein the microarray is selected from the group consisting of: HuGeneFL Array (Affymetrix), Human Gene Cancer Gl 10 Array (Affymetrix), Human Genome Focus Array (Affymetrix), Human Genome Ul 33 Set (Affymetrix), Human Genome U95 Set (Affymetrix), Human IA Oligo Microarray Kit (V2) (Agilent Technologies), Custom Gene Expression Microarray (Agilent Technologies), CatalogArray (CombiMatrix), CodeLink UniSet® Human 20KI (GE Healthcare), HumanTox-16 BeadChip (Illumina), HumanTox-96 Array Matrix (Illumina), and Custom OciChip arrays (Ocimum Biosolutions).

15. The method of any of claims 10-14, further comprising the step of obtaining the microarray.

16. The method of claim 1, wherein the RNA species correspond to nucleic acid molecules encoding a functional peptide selected from the group consisting of cytokines and antioxidants.

17. The method of claim 1, wherein the patient is experiencing trauma.

18. The method of claim 1, wherein the patient is in shock.

19. The method of claim 1 , wherein the patient has a burn.

20. The method of claim 1, wherein the biological sample comprises blood cells.

21. The method of claim 20, wherein the blood cells comprise white blood cells.

22. A kit for assessing prognosis of a patient comprising a computer readable medium comprising: (i) information corresponding to one or more RNA expression profiles of a biological sample from the patient and RNA expression profiles of a reference sample; and (ii) RNA expression profile analysis software capable of being loaded into the memory of a computer system.

23. The kit of claim 22, wherein the computer readable medium is a microarray.

24. The kit of claim 22, further comprising instructions for use.

25. A system for assessing prognosis of a patient, comprising; a) means for measuring the expression level of a plurality of RNA species in a biological sample obtained from the patient; b) means for determining a comparative value for each RNA species by comparing the expression level for each mRNA species with a reference value; c) means for computing a score indicative of the patient's condition using the comparative values; and d) means for displaying the score to a user to facilitate assessment of a prognosis for the patient.

26. The system of claim 25, wherein the means for computing a score computes the score at least in part by performing a summation operation using the comparative values or values derived from the comparative values.

27. The system of claim 26, wherein the means for computing a score computes the score by performing at least one mathematical operation on the comparative values prior to performing the summation operation.

28. The system of claim 27, wherein the mathematical operation is chosen from the group consisting of: a) multiplying the comparative value by itself at least once; and b) multiplying the comparative value by a product of a standard deviation relating to the reference value.

29. The system of claim 28, wherein the product of the standard deviation is an inverse of a square of the standard deviation.

30. The system of claim 25, wherein the score is displayed on a portable electronic device.

31. The system of claim 25, wherein the means for determining a comparative value is adapted and configured to access a database having a plurality of reference values stored thereon.

32. The system of claim 31 , wherein the database is accessed by the system over the internet.

33. The system of claim 25, wherein the means for determining a comparative value computes at least one comparative value by: a) subtracting the reference value from the expression level to create a first intermediate value; b) squaring the first intermediate value to create a second intermediate value; and c) dividing the second intermediate value by the square of a standard deviation relating to the reference value to create a third intermediate value.

34. The system of claim 25, wherein the system is adapted and configured to compute a score indicative of a patient's condition for a population of patients.

35. The system of claim 34, further comprising means for prioritizing treatment for each of the patients in the population based on the score of each patient.

36. The system of claim 35, wherein a patient with a higher score is given treatment priority over a patient with a lower score.

37. The system of claim 25, wherein the means for measuring the expression level includes at least one microarray.

38. The system of claim 37, wherein the at least one microarray includes a whole genome expression array.

39. The system of claim 38, wherein the microarray is selected from the group consisting of: Human Genome Survey Microarray v2.0 (Applied Biosystems), Human Genome U133 Plus 2.0 Array (Affymetrix), Gene Chip® Human X3P Array (Affymetrix), Whole Human Genome Oligo Microarray Kit (Agilent Technologies), CodeLink® Human Whole Genome Bioarray (GE Healthcare), Sentrix® Human-6 Expression BeadChip (Illumina), Sentrix® HumanRef-8 Expression BeadChip (Illumina), and Human 4OK OciChip™ (Ocimum Biosolutions).

40. The system of claim 37, wherein the at least one microarray includes a selected expression array.

41. The system of claim 40, wherein the microarray is selected from the group consisting of: HuGeneFL Array (Affymetrix), Human Gene Cancer Gl 10 Array (Affymetrix), Human Genome Focus Array (Affymetrix), Human Genome Ul 33 Set (Affymetrix), Human Genome U95 Set (Affymetrix), Human IA Oligo Microarray Kit (V2) (Agilent Technologies), Custom Gene Expression Microarray (Agilent Technologies), CatalogArray (CombiMatrix), CodeLink UniSet® Human 20KI (GE Healthcare), HumanTox-16 BeadChip (Illumina), HumanTox-96 Array Matrix (Illumina), and Custom OciChip arrays (Ocimum Biosolutions).

42. A machine readable program, tangibly embodied on a computer readable medium, containing instructions for controlling a system for predicting clinical outcome of a patient comprising: a) means for determining a comparative value for each of a plurality of RNA species extracted from a patient to be assessed by comparing an expression level for each RNA species with an associated reference value; and b) means for computing a score indicative of the patient's condition using the comparative values to facilitate assessment of a prognosis for the patient.

43. The machine readable program of claim 42, wherein the means for computing a score computes the score at least in part by performing a summation operation using the comparative values or values derived from the comparative values.

44. The machine readable program of claim 43, wherein the means for computing a score computes the score by performing at least one mathematical operation on the comparative values prior to performing the summation operation.

45. The machine readable program of claim 44, wherein the mathematical operation is chosen from the group consisting of: a) multiplying the comparative value by itself at least once; and b) multiplying the comparative value by a product of a standard deviation relating to the reference value.

46. The machine readable program of claim 45, wherein the product of the standard deviation is an inverse of a square of the standard deviation.

47. The machine readable program of claim 42, wherein the means for determining a comparative value is adapted and configured to access a database having a plurality of reference values stored thereon.

48. The machine readable program of claim 47, wherein the means for determining a comparative value is adapted and configured to access the database through the internet.

49. The machine readable program of claim 42, wherein the means for determining a reference value computes at least one reference value by: a) subtracting the reference value from the expression level to create a first intermediate value; b) squaring the first intermediate value to create a second intermediate value; and c) dividing the second intermediate value by the square of a standard deviation relating to the reference value to create a third intermediate value.

50. The machine readable program of claim 42, wherein the program includes means for computing a score indicative of a patient's condition for a population of patients.

51. The machine readable program of claim 50, wherein the program further includes means for prioritizing treatment to determine treatment priority for each of the patients in the population based at least in part on the score of each patient.

52. The machine readable program of claim 51 , wherein the means for prioritizing treatment also determines treatment priority based on the particular medical condition of the patient.

53. The machine readable program of claim 52, wherein the means for prioritizing treatment also determines treatment priority based on the severity of the medical condition of the patient.

54. The machine readable program of claim 53, wherein a patient with a higher score is assigned treatment priority in a queue over a patient with a lower score by the means for prioritizing treatment.

55. The machine readable program of claim 54, further comprising means for periodically updating the queue.

56. The machine readable program of claim 42, wherein the program is adapted and configured to display the score on a portable electronic device.

57. The machine readable program of claim 54, wherein the program is adapted and configured to display the queue on a portable electronic device.

58. The machine readable program of claim 42, wherein the program further includes means for assigning a treatment priority category to a patient based at least in part on the patient's score.

59. The machine readable program of claim 58, wherein the means for assigning a treatment priority category is also based on the particular medical condition of the patient.

60. The machine readable program of claim 51 , wherein the means for assigning a treatment priority category determines treatment priority based on the severity of the medical condition of the patient.