CN110993104B - Tumor patient lifetime prediction system - Google Patents
Tumor patient lifetime prediction system Download PDFInfo
- Publication number
- CN110993104B CN110993104B CN201911222137.0A CN201911222137A CN110993104B CN 110993104 B CN110993104 B CN 110993104B CN 201911222137 A CN201911222137 A CN 201911222137A CN 110993104 B CN110993104 B CN 110993104B
- Authority
- CN
- China
- Prior art keywords
- risk score
- gene
- relative expression
- tumor
- survival
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The present disclosure relates to a tumor patient survival prediction system, the system comprising: the tumor prognosis related gene expression system comprises a computing device, an input device for inputting a risk score of a tumor prognosis related gene pair relative expression quantity of an individual tumor patient and an output device for outputting the survival probability of the survival period of the tumor patient; wherein the tumor prognosis related gene pair is a gene pair with the relative size of the expression quantity of two genes in the gene pair and the survival time of a tumor patient; the computing device includes a memory and a processor; the memory stores computer programs to realize modeling algorithm and algorithm of discriminant function; the modeling algorithm is a least squares algorithm. The survival time prediction system for the tumor patients integrates clinically easily available variables, can rapidly and accurately obtain the prediction result of the survival time of the tumor patients, and saves time and energy of users.
Description
Technical Field
The present disclosure relates to computer application technology, and in particular, to a tumor patient survival prediction system.
Background
The method has important clinical, scientific and social values for accurately predicting the life cycle of the patient. In clinical work, accurate life prediction can guide doctors to make personalized examination and treatment schemes for high-risk patients, help doctors to make reasonable review and follow-up plans, and further improve the quality of medical service. In scientific research, accurate prediction of patient risk levels can provide important basis for developing effective treatment schemes for high-risk patients, and can become an important reference for checking novel treatment effects. From the social aspect, the survival time of the patient can be accurately predicted, scientific survival expectation can be provided for the patient and family members, the patient is guided to follow the treatment plan, excessive medical treatment is avoided, the economic pressure of families is lightened, and the doctor-patient relationship is improved.
Currently, existing methods for predicting survival of tumor patients mainly rely on molecular markers and gene expression levels. However, the molecular marker has the problems of overfitting, too small number of found group cases, lack of external verification and the like, and cannot be applied to clinical practice; the problem that heterogeneity among data sets, cross-platform detection technical deviation and the like exist in predicting the survival time of a patient by using the gene expression quantity, and clinical application is difficult to realize. Therefore, there is a need for a simple, reliable and highly accurate prediction system for survival of tumor patients.
Disclosure of Invention
The purpose of the present disclosure is to provide a tumor patient lifetime prediction system, with which the lifetime of a tumor patient can be predicted simply, reliably and accurately.
In order to achieve the above object, the present disclosure provides a tumor patient survival prediction system including a computing device, an input device for inputting a risk score of a tumor prognosis-related gene pair relative expression amount of an individual tumor patient, and an output device for outputting a survival probability of a tumor patient;
wherein the tumor prognosis related gene pair is a gene pair with the relative size of the expression quantity of two genes in the gene pair and the survival time of a tumor patient;
the computing device includes a memory and a processor; the memory stores a computer program to realize a modeling algorithm and an algorithm of a discriminant function shown in the formula (1); the modeling algorithm is a least partial square algorithm;
F(x，y)＝f(x 1 ) F (y) formula (1),
wherein x is 1 A risk score representing the relative expression level of each pair of the tumor prognosis-related genes, y representing the survival of the tumor patient, f (x 1 ) And (3) representing the total risk score of the tumor prognosis related gene to the relative expression quantity, F (y) representing a survival rate curve function with the survival time of y years, and F (x, y) representing the probability of the survival time of y years of the tumor patient.
Optionally, the calculation method of the risk score of the tumor prognosis related gene to the relative expression quantity is as follows:
in any of the tumor prognosis-related gene pairs, if the expression level of the first gene is greater than the expression level of the second gene, the relative expression level risk score x of the gene pair 1i Risk score for the first relative expression quantity; if the expression level of the first gene is less than or equal to the expression level of the second gene, the relative expression level risk score x of the gene pair 1i Risk score for the second relative expression quantity; wherein x is 1i Refers to the risk score of the i-th relative expression level of the tumor prognosis related gene pair.
Optionally, the value of the first relative expression risk score is greater than 0, and the value of the second relative expression risk score is less than or equal to 0.
Optionally, the input device is further configured to input a 1p19q status risk score and a pathology level risk score for the individual tumor patient;
the computing means is also for implementing an algorithm of a discriminant function as shown in equation (2),
F(x，y)＝[f(x 1 )+f(x 2 )+f(x 3 )]f (y) formula (2),
wherein x is 1 A risk score, x, representing the relative expression level of each pair of said tumor prognosis-related genes 2 Representing the 1p19q state risk score, x 3 Representing the pathological grade risk score, y representing the survival of the tumor patient, f (x) 1 )+f(x 2 )+f(x 3 ) Representing the total risk score, F (y) representing a survival curve function with a survival of y years, and F (x, y) representing the probability of the tumor patient with a survival of y years.
Optionally, the method for calculating the 1p19q state risk score includes:
when the 1p19q chromosome is in a deletion state, the 1p19q state risk score x 2 A first 1p19q state risk score; when the 1p19q chromosome is in a complete state, the 1p19q state risk score x 2 A second 1p19q state risk score;
the method for calculating the pathological level risk score comprises the following steps:
at a pathology level of WHO class II, the pathology level risk score x 3 A first pathology level risk score; when the pathology level is WHO III level, the pathology level risk score x 3 And a second pathology level risk score.
Optionally, the value of the first 1p19q state risk score is greater than 0, and the value of the second 1p19q state risk score is less than or equal to 0;
the value of the first pathological level risk score is smaller than or equal to 0, and the value of the second pathological level risk score is larger than 0.
Alternatively, the process may be carried out in a single-stage,
wherein a is i A risk score coefficient indicating the relative expression amount of the ith gene pair related to tumor prognosis, n being a positive integer of 1 or more;
f(x 2 )＝bx 2 wherein b represents a 1p19q state risk score coefficient;
f(x 3 )＝cx 3 wherein c represents a pathology level risk score coefficient.
Optionally, the tumor patient comprises at least one of glioma patient, pancreatic cancer, prostate cancer, nasopharyngeal cancer, endometrial cancer, thyroid cancer, liver cancer, breast cancer, colorectal tumor, bladder cancer, gastric cancer, lung cancer, melanoma, renal clear cell tumor, and adrenocortical cancer.
Optionally, the calculating device can establish a survival probability alignment chart and calculate a total risk score based on the risk score of the tumor prognosis related gene input by the input device on the relative expression amount, the 1p19q state risk score and the pathology level risk score, wherein the total risk score is the sum of the risk score of the tumor prognosis related gene input by the input device on the relative expression amount, the 1p19q state risk score and the pathology level risk score, and calculates a survival predicted value of the tumor patient according to the total risk score;
the method for establishing the survival probability alignment chart based on the risk scores of the tumor prognosis related genes, the 1p19q state risk scores and the pathology level risk scores input by the input device is to apply an R language RMS operation package to complete alignment chart visualization of a Cox regression model.
Optionally, the system further comprises a detection device for the relative expression quantity of the tumor prognosis related gene pair, and the detection device comprises a detection chip and a chip signal reader for the relative expression quantity of the tumor prognosis related gene pair.
Through the technical scheme, the tumor patient survival prediction system predicts the survival of the tumor patient based on the relative expression quantity of the tumor prognosis related gene pair of the individual tumor patient, the variable is easy to obtain clinically, and the prediction result of the survival of the tumor patient can be obtained rapidly and accurately by utilizing the variable, so that the time and energy of a user are saved.
Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings are included to provide a further understanding of the disclosure, and are incorporated in and constitute a part of this specification, illustrate the disclosure and together with the description serve to explain, but do not limit the disclosure. In the drawings:
fig. 1 is a schematic diagram of a tumor patient survival prediction system provided by an embodiment of the present disclosure.
Fig. 2 is a lower-level glioma patient survival probability alignment graph provided by an embodiment of the present disclosure.
Detailed Description
Specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the disclosure, are not intended to limit the disclosure.
A first aspect of the present disclosure provides a tumor patient survival prediction system, as shown in fig. 1, comprising a computing device, an input device for inputting a risk score of a tumor prognosis related gene versus a relative expression amount of an individual tumor patient, and an output device for outputting a survival probability of a tumor patient; wherein the tumor prognosis related gene pair is a gene pair with the relative size of the expression quantity of two genes in the gene pair and the survival time of a tumor patient; the computing device includes a memory and a processor; the memory stores a computer program to realize a modeling algorithm and an algorithm of a discriminant function shown in the formula (1); the modeling algorithm is a least partial square algorithm;
F(x，y)＝f(x 1 ) F (y) formula (1),
wherein x is 1 A risk score representing the relative expression level of each pair of the tumor prognosis-related genes, y representing the survival of the tumor patient, f (x 1 ) And (3) representing the total risk score of the tumor prognosis related gene to the relative expression quantity, F (y) representing a survival rate curve function with the survival time of y years, and F (x, y) representing the probability of the survival time of y years of the tumor patient.
The tumor patient survival prediction system provided by the disclosure predicts the survival of a tumor patient based on the relative expression quantity of the tumor prognosis related gene pair of the individual tumor patient, the variable is easy to obtain clinically, and the prediction result of the survival of the tumor patient can be obtained rapidly and accurately by using the variable, so that the time and energy of a user are saved.
According to the present disclosure, the risk score of the tumor prognosis-related gene for the relative expression amount can be calculated by the following method:
in any of the tumor prognosis-related gene pairs, if the expression level of the first gene is greater than the expression level of the second gene, the relative expression level risk score x of the gene pair 1i Risk score for the first relative expression quantity; if the expression level of the first gene is less than or equal to the expression level of the second gene, the relative expression level risk score x of the gene pair 1i Risk score for the second relative expression quantity; wherein x is 1i Refers to the risk score of the i-th relative expression level of the tumor prognosis related gene pair. Wherein the relative expression levels of the constitutive genes of any one of the tumor prognosis-related gene pairs can be obtained by a conventional method in the artThe expression level of the constituent genes of any of the tumor prognosis-related gene pairs can be obtained by, for example, second-generation sequencing, gene chip or PCR method.
The values of the first relative expression risk score and the second relative expression risk score can be changed in a larger range, and the values of the first relative expression risk score and the second relative expression risk score can be set according to the influence of the expression products of the constituent genes of each gene pair on the survival time of the patient. For example, the value of the first relative expression risk score may be greater than 0, and the value of the second relative expression risk score may be equal to or less than 0. As a preferred aspect of the present disclosure, the first relative expression risk score may have a value of 1 and the second relative expression risk score may have a value of 0.
In accordance with the present disclosure, to increase the accuracy of the prediction of the survival of a tumor patient, the input device may also be used to input a 1p19q status risk score and a pathology level risk score of an individual tumor patient, as shown in fig. 2;
the computing means is also for implementing an algorithm of a discriminant function as shown in equation (2),
F(x，y)＝[f(x 1 )+f(x 2 )+f(x 3 )]f (y) formula (2),
wherein x is 1 A risk score, x, representing the relative expression level of each pair of said tumor prognosis-related genes 2 Representing the 1p19q state risk score, x 3 Representing the pathological grade risk score, y representing the survival of the tumor patient, f (x) 1 )+f(x 2 )+f(x 3 ) Representing the total risk score, F (y) representing a survival curve function with a survival of y years, and F (x, y) representing the probability of the tumor patient with a survival of y years.
Through the technical scheme, the tumor patient survival prediction system provided by the disclosure integrates clinically easily obtained variables, and the prediction result of the survival of the lower-level glioma patient can be obtained rapidly and accurately by inputting the variables.
According to the present disclosure, the 1p19q state risk score may be calculated by:
when the 1p19q chromosome is in a deletion state, the 1p19q state risk score x 2 A first 1p19q state risk score; when the 1p19q chromosome is in a complete state, the 1p19q state risk score x 2 And the second 1p19q state risk score. The method for obtaining the 1p19q chromosome association deletion state may be a conventional method in the art, and the present disclosure is not limited, for example, the 1p19q chromosome association deletion state may be detected using a fluorescence in situ hybridization method.
The values of the first 1p19q state risk score and the second 1p19q state risk score may vary within a larger range, and the values of the first 1p19q state risk score and the second 1p19q state risk score may be set according to the influence of the 1p19q state on the survival time of the patient. For example, the value of the first 1p19q state risk score may be greater than 0, and the value of the second 1p19q state risk score may be equal to or less than 0. As a preferred aspect of the present disclosure, the first 1p19q state risk score may have a value of 1 and the second 1p19q state risk score may have a value of 0.
According to the present disclosure, the pathology level risk score may be calculated by:
at a pathology level of WHO class II, the pathology level risk score x 3 A first pathology level risk score; when the pathology level is WHO III level, the pathology level risk score x 3 And a second pathology level risk score. Wherein the pathology level is a glioma pathology level determined according to the World Health Organization (WHO) hierarchy system.
The values of the first pathological level risk score and the second pathological level risk score can be changed in a larger range, and specific design can be performed according to the influence of the pathological level on the survival time of the patient. For example, the first pathology level risk score may have a value of 0 or less, and the second pathology level risk score may have a value of 0 or more. As a preferred aspect of the present disclosure, the first pathology level risk score may have a value of 0 and the second pathology level risk score may have a value of 1.
Specifically, in the above technical solution, a risk score greater than 0 indicates that the patient lifetime is affected, and a risk score less than or equal to 0 indicates that the patient lifetime is not affected or is helpful to extend the patient lifetime.
According to the present disclosure, the method of calculating the sum of the accumulated risk scores of the tumor prognosis-related genes corresponding to the relative expression amounts, the 1p19q states, and the pathology levels, respectively, to obtain the total risk score may be a conventional method in the art. For example, the risk scores corresponding to the relative expression amounts, the 1p19q states and the pathology levels of the tumor prognosis-related gene pairs can be directly added to obtain total risk scores; the risk scores corresponding to the relative expression quantity, the 1p19q state and the pathology level of the tumor prognosis related genes can be weighted and added to obtain the total risk score.
In a preferred embodiment of the present invention,
wherein a is i A risk score coefficient indicating the relative expression amount of the ith gene pair related to tumor prognosis, n being a positive integer of 1 or more;
f(x 2 )＝bx 2 wherein b represents a 1p19q state risk score coefficient;
f(x 3 )＝cx 3 wherein c represents a pathology level risk score coefficient. In the above preferred case, the risk score coefficient can reflect the influence degree of each variable on the life cycle of the patient, and can improve the prediction accuracy of the prediction system of the present disclosure. The extent of influence of the individual variables on the patient lifetime can be determined by LASSO regression calculation, so as to determine the values of a, b and c, and LASSO regression analysis is a conventional technical means in the field, and the disclosure is not repeated here.
According to the present disclosure, the tumor patient may be selected within a broad scope, for example, the tumor patient includes at least one of glioma patient, pancreatic cancer, prostate cancer, nasopharyngeal cancer, endometrial cancer, thyroid cancer, liver cancer, breast cancer, colorectal tumor, bladder cancer, gastric cancer, lung cancer, melanoma, renal clear cell tumor, and adrenocortical cancer.
Illustratively, the tumor patient may be a lower grade glioma patient, and the World Health Organization (WHO) classification system classifies gliomas into grade i to grade iv based on histological features, with recent studies describing grade ii and grade iii gliomas as lower grade gliomas. In this exemplary embodiment, the tumor prognosis related gene pairs may include CRH-IFNB1, HOXA9-PRG3, IL10-IL9, IL9-PTH2, IL9-RETNLB, NKX2-5-PRLH, NKX3-2-UCN3, NR2C1-PTX3, PRLHR-REG1A, and PRLHR-TRIM31.
Wherein, the reference number of the CRH gene in the NCBI database is 1392; the IFNB1 gene is numbered 3456 in the NCBI database; the HOXA9 gene is numbered 3205 in the NCBI database; the PRG3 gene is referenced 10394 in the NCBI database; the IL10 gene is referenced 3586 in NCBI database; the IL9 gene is referenced 3578 in NCBI database; RETNLB gene has a reference number 84666 in NCBI database; the NKX2-5 gene is referenced 1482 in the NCBI database; the PRLH gene has the reference number 51052 in the NCBI database; the NKX3-2 gene has the reference number 579 in the NCBI database; the UCN3 gene has a reference number 114131 in the NCBI database; the NR2C1 gene is referenced 7181 in the NCBI database; the PTX3 gene has a reference number of 5806 in NCBI database; the PRLHR gene is referenced 2834 in the NCBI database; the REG1A gene has a reference number 5967 in NCBI database; the TRIM31 gene has a reference number 11074 in the NCBI database. The gene pairs are immune gene pairs, and through the expression products of the immune gene pairs, the immune state of a patient can be estimated, wherein the immune state is an important cause of malignant development of glioma, and accurate estimation of the immune response state is beneficial to improving the accuracy of life prediction of the patient.
The method for calculating the relative expression quantity of the lower-grade glioma prognosis-related immune gene pair comprises the following steps:
if the expression level of the CRH gene is greater than the expression level of the IFNB1 gene, a first relative expression level of the CRH-IFNB1 gene pair is greater than that of the first relative expression levelRisk score x 11 1 is shown in the specification; if the expression level of the CRH gene is less than or equal to the expression level of the IFNB1 gene, a second relative expression level risk score x of the CRH-IFNB1 gene pair 11 Is 0;
if the expression level of the HOXA9 gene is greater than the expression level of the PRG3 gene, a first relative expression level risk score x of the HOXA9-PRG3 gene pair 12 1 is shown in the specification; if the expression level of the HOXA9 gene is less than or equal to the expression level of the PRG3 gene, a second relative expression level risk score x of the HOXA9-PRG3 gene pair 12 Is 0;
if the expression level of the IL10 gene is greater than the expression level of the IL9 gene, a first relative expression level risk score x for the IL10-IL9 gene pair 13 1 is shown in the specification; if the expression level of the IL10 gene is less than or equal to the expression level of the IL9 gene, a second relative expression level risk score x of the IL10-IL9 gene pair 13 Is 0;
if the expression level of the IL9 gene is greater than the expression level of the PTH2 gene, a first relative expression level risk score x of the IL9-PTH2 gene pair 14 1 is shown in the specification; if the expression level of the IL9 gene is less than or equal to the expression level of the PTH2 gene, a second relative expression level risk score x of the IL9-PTH2 gene pair 14 Is 0;
if the expression level of the IL9 gene is greater than the expression level of the RETNLB gene, a first relative expression level risk score x of the IL9-RETNLB gene pair 15 1 is shown in the specification; if the expression level of the IL9 gene is less than or equal to the expression level of the RETNLB gene, a second relative expression level risk score x of the IL9-RETNLB gene pair 15 Is 0;
if the expression level of the NKX2-5 gene is greater than the expression level of the PRLH gene, a first relative expression level risk score x of the NKX2-5-PRLH gene pair 16 1 is shown in the specification; if the expression level of the NKX2-5 gene is less than or equal to the expression level of the PRLH gene, a second relative expression level risk score x of the NKX2-5-PRLH gene pair 16 Is 0;
if the expression level of the NKX3-2 gene is greater than that of the UCN3 geneThe first relative expression risk score x of the NKX3-2-UCN3 gene pair due to the expression level 17 1 is shown in the specification; if the expression level of the NKX3-2 gene is less than or equal to the expression level of the UCN3 gene, a second relative expression level risk score x of the NKX3-2-UCN3 gene pair 17 Is 0;
if the expression level of the NR2C1 gene is greater than the expression level of the PTX3 gene, a first relative expression level risk score x of the NR2C1-PTX3 gene pair 18 1 is shown in the specification; if the expression level of the NR2C1 gene is less than or equal to the expression level of the PTX3 gene, a second relative expression level risk score x of the NR2C1-PTX3 gene pair 18 Is 0;
if the expression level of the PRLHR gene is greater than the expression level of the REG1A gene, a first relative expression level risk score x of the PRLHR-REG1A gene pair 19 1 is shown in the specification; if the expression level of the PRLHR gene is less than or equal to the expression level of the REG1A gene, a second relative expression level risk score x of the PRLHR-REG1A gene pair 19 Is 0;
if the expression level of the PRLHR gene is greater than the expression level of the TRIM31 gene, a first relative expression level risk score x for the PRLHR-TRIM31 gene pair 110 1 is shown in the specification; if the expression level of the PRLHR gene is less than or equal to the expression level of the TRIM31 gene, a second relative expression level risk score x of the PRLHR-TRIM31 gene pair 110 Is 0.
Wherein, the relative expression amount risk score coefficient a of the gene to CRH-IFNB1 1 Risk score coefficient a for relative expression level of HOXA9-PRG3 of-0.00332 gene 2 Risk score coefficient a for relative expression level of 0.124745 and gene to IL10-IL9 3 Risk score coefficient a for relative expression level of 0.205056 gene to IL9-PTH2 4 Risk score coefficient a for relative expression level of-0.11688 gene to IL9-RETNLB 5 Risk score coefficient a of relative expression level of-0.30775 gene to NKX2-5-PRLH 6 Risk score coefficient a for relative expression level of 0.035793 gene to NKX3-2-UCN3 7 Risk score coefficient a for relative expression level of 0.086457, gene pair NR2C1-PTX3 8 Is-0.49552. Relative expression level risk score coefficient a of gene pair PRLHR-REG1A 9 Risk score coefficient a of relative expression amount of-0.24355 and gene pair PRLHR-TRIM31 10 Is-0.33529;
the 1p19q state risk score coefficient b is-1.1591;
the risk score coefficient c of the pathology level is 1.0261.
In the above-described exemplary embodiment, the absolute value of the risk score coefficient indicates the degree of influence of each variable on the patient's lifetime, and the greater the absolute value, the greater the degree of influence; and the risk score coefficient is positive or negative, the influence mode of each variable on the life cycle of the patient is represented, when the risk score coefficient is positive, the variable is represented as a destructive factor, the life cycle of the patient is shortened, and when the risk score coefficient is negative, the variable is represented as a protective factor, and the life cycle of the patient is prolonged. In this preferred manner, the prediction results of the prediction system of the present disclosure are made more accurate.
As another exemplary embodiment of the present application, the tumor patient may be a pancreatic cancer patient, and in this exemplary embodiment, the tumor prognosis-related gene pair may include: LEMD2-MYD88, DNMT3A-MECOM, AXIN1-RRAS2, BRCA1-GPS2, CCND1-KEAP1, NOTCH2-TBX3, MET-SPTAN1. The genes involved in the gene pairs are all internationally recognized tumor initiating genes.
Wherein, the LEMD2 gene has a reference number 221496 in NCBI database; the MYD88 gene is referenced 4615 in NCBI database; the DNMT3A gene has the reference number 1788 in the NCBI database; the reference number of the MECOM gene in the NCBI database is 2122; the AXIN1 gene has the reference number 8312 in the NCBI database; the RRAS2 gene has a reference number 22800 in the NCBI database; the BRCA1 gene is numbered 672 in the NCBI database; the GPS2 gene has the reference number 2874 in NCBI database; the CCND1 gene is referenced 595 in the NCBI database; the KEAP1 gene has a reference number 9817 in the NCBI database; the NOTCH2 gene is numbered 4853 in the NCBI database; the TBX3 gene has a reference number 6926 in the NCBI database; the MET gene is referenced 4233 in the NCBI database; the SPTAN1 gene has the reference number 6709 in the NCBI database.
The method for calculating the relative expression quantity of the pancreatic cancer prognosis-related gene pair comprises the following steps:
if the expression level of the LEMD2 gene is greater than the expression level of the MYD88 gene, a first relative expression level risk score x of the LEMD2-MYD88 gene pair 11 1 is shown in the specification; if the expression level of the LEMD2 gene is less than or equal to the expression level of the MYD88 gene, a second relative expression level risk score x of the LEMD2-MYD88 gene pair 11 Is 0;
if the expression level of the DNMT3A gene is greater than the expression level of the MECOM gene, a first relative expression level risk score x of the DNMT3A-MECOM gene pair 12 1 is shown in the specification; if the expression level of the DNMT3A gene is less than or equal to the expression level of the MECOM gene, a second relative expression level risk score x of the DNMT3A-MECOM gene pair 12 Is 0;
if the expression level of the AXIN1 gene is greater than the expression level of the RRAS2 gene, a first relative expression level risk score x of the AXIN1-RRAS2 gene pair 13 1 is shown in the specification; if the expression level of the AXIN1 gene is less than or equal to the expression level of the RRAS2 gene, a second relative expression level risk score x of the AXIN1-RRAS2 gene pair 13 Is 0;
if the expression level of the BRCA1 gene is greater than the expression level of the GPS2 gene, a first relative expression level risk score x of the BRCA1-GPS2 gene pair 14 1 is shown in the specification; if the expression level of the BRCA1 gene is less than or equal to the expression level of the GPS2 gene, a second relative expression level risk score x of the BRCA1-GPS2 gene pair 14 Is 0;
if the expression level of the CCND1 gene is greater than the expression level of the KEAP1 gene, a first relative expression level risk score x of the CCND1-KEAP1 gene pair 15 1 is shown in the specification; if the expression level of the CCND1 gene is less than or equal to the expression level of the KEAP1 gene, a second relative expression level risk score x of the CCND1-KEAP1 gene pair 15 Is 0;
if the expression level of the NOTCH2 gene isGreater than the expression level of the TBX3 gene, then the first relative expression level risk score x of the NOTCH2-TBX3 gene pair 16 1 is shown in the specification; if the amount of expression of the NOTCH2 gene is less than or equal to the amount of expression of the TBX3 gene, a second relative amount of expression risk score x for the NOTCH2-TBX3 gene pair 16 Is 0;
if the expression level of the MET gene is greater than the expression level of the SPTAN1 gene, a first relative expression level risk score x of the MET-SPTAN1 gene pair 17 1 is shown in the specification; if the expression level of the MET gene is less than or equal to the expression level of the SPTAN1 gene, a second relative expression level risk score x of the MET-SPTAN1 gene pair 17 Is 0.
Wherein, the relative expression amount risk score coefficient a of the gene pair LEMD2-MYD88 1 Risk score coefficient a of relative expression quantity of gene pair DNMT3A-MECOM of-0.2385 2 Relative expression level risk score coefficient a of-0.0308 for gene pair AXIN1-RRAS2 3 Risk score coefficient a of relative expression quantity of-0.02498 and gene to BRCA1-GPS2 4 Risk score coefficient a for relative expression level of 0.009342 and gene pair CCND1-KEAP1 5 Risk score coefficient a for relative expression level of 0.189372, gene to NOTCH2-TBX3 6 Risk score coefficient a for relative expression amount of 0.194922 and gene to MET-SPTAN1 7 0.390684.
According to the disclosure, the computing device can establish a survival probability alignment chart and calculate a total risk score based on the risk score of the tumor prognosis related gene input by the input device on the relative expression amount, the 1p19q state risk score and the pathology level risk score, wherein the total risk score is the sum of the risk score of the tumor prognosis related gene input by the input device on the relative expression amount, the 1p19q state risk score and the pathology level risk score, and calculate a survival predicted value of a tumor patient according to the total risk score;
the method for establishing the survival probability alignment chart based on the risk scores of the tumor prognosis related genes, the 1p19q state risk scores and the pathology level risk scores input by the input device is to apply an R language RMS operation package to complete alignment chart visualization of a Cox regression model.
Fig. 2 is a lower-level glioma patient survival probability alignment graph provided by an embodiment of the present disclosure. As shown in fig. 2, in the survival probability alignment, the relative expression amount of the tumor prognosis-related gene, the 1p19q state, and the pathology level correspond to different risk score ranges, respectively. Calculating the accumulated sum of the risk scores corresponding to the 3 variables as a total risk score, and drawing a vertical line of a 1-year survival rate line of the patient at the position of the total risk score in the survival probability alignment chart, wherein the intersection point of the vertical line and the 1-year survival rate line of the patient is the probability that the survival period of the patient is 1 year; the probability of a patient survival of 2 years, 3 years or 5 years can be obtained using the same method. Different total risk scores correspond to probabilities of different patient survival being 1 year, 2 years, 3 years, and 5 years.
According to the present disclosure, the system may further include a detection device for the relative expression amount of the tumor prognosis related gene pair, and the detection device includes a detection chip and a chip signal reader for the relative expression amount of the tumor prognosis related gene pair.
According to the present disclosure, the input device and the computing device may be connected by a wired manner and/or a wireless manner; the computing device and/or the output device may be connected by a wired and/or wireless means. The wireless connection mode can be wireless local area network, bluetooth, infrared and the like; the wired connection mode can be USB, fixed telephone network, etc. The adoption of the connection mode can greatly facilitate the use of a prediction system by a user, and meanwhile, the life time of a tumor patient can be accurately predicted by means of increasingly developed information technology and increasingly popular network resources.
According to the disclosure, the input device may be a keyboard, a mouse, a touch screen, or the like; the output device may be a display, a printer or an audio output device; the computing device may be a host computer, a central processing unit, or a web server.
The present disclosure is further illustrated by the following examples, but the present disclosure is not limited thereby.
Example 1
The embodiment takes lower-grade glioma as an example, and is used for explaining the establishment method of the tumor patient survival prediction system model
A group of 172 lower-grade glioma patient transcriptome expression profile data from China is collected as a discovery group, 2214 immunity genes related to the lower-grade glioma are screened out from the discovery group data, and the expression quantity of the immunity genes is compared pairwise to establish a database containing 2449791 immunity gene pairs. Assuming that the i gene and the j gene are a pair of genes in the database, in one patient, if the expression level of the i gene is greater than the expression level of the j gene, the pair of genes is denoted as 1, and if the expression level of the i gene is equal to or less than the expression level of the j gene, the pair of genes is denoted as 0. If the expression results of a gene pair in more than 95% of patients are recorded as 1 or 0, the gene pair is deleted from the database, and finally 15957 gene pairs in the database are retained and subjected to subsequent analysis.
Through the log-rank test, it is known that 4464, 4911 and 3557 immune gene pairs in the discovery group, the internal verification group and the external verification group respectively have obvious prognosis values, and 402 immune gene pairs with common prognosis values are obtained after intersection of the three immune gene pairs.
LASSO regression analysis was performed on the 402 immune gene pairs with general prognostic value to determine 10 gene pairs with significant prognostic value, as follows: CRH-IFNB1, HOXA9-PRG3, IL10-IL9, IL9-PTH2, IL9-RETNLB, NKX2-5-PRLH, NKX3-2-UCN3, NR2C1-PTX3, PRLHR-REG1A and PRLHR-TRIM31, and the relative expression amounts of the 10 genes with remarkable prognostic value to the expression products in the patient were obtained.
And (3) carrying out joint analysis on the relative expression quantity of the prognosis related gene pair and clinical common molecular characteristics such as age, pathological grade, 1p19q state, IDH mutant filling and the like through a COX regression model, and finally determining that the relative expression quantity of the prognosis related gene pair, the 1p19q state and the pathological grade are independently related to the survival time of a patient. Based on the relative expression quantity, the 1p19q state and the pathological level of the prognosis related gene, a Cox regression model of a lower-level glioma patient is established, the line graph is subjected to visual transformation, and a survival probability line graph of the lower-level glioma patient is established, wherein the survival probability line graph is shown in fig. 2. The specific process of establishing the survival probability nomograms of lower-level glioma patients is to apply an R language RMS operation package to convert the obtained Cox regression model into a visualized survival probability nomogram. The specific commands are as follows:
nom </nomogram (COX regression model,
fun＝list(function(x)surv(12,x),
function(x)surv(24,x),
function(x)surv(36,x),
function(x)surv(60,x)),
lp＝F,
funlabel＝c("1-year survival",
2-year survival",
"3-year survival",
"5-year survival"),
fun.at＝seq(0.9,0,by＝-0.1))
plot(nom,cex.axis＝1,fontsize1＝0.5)。
through the steps, a survival probability alignment chart of the glioma patient with the lower grade shown in the figure 2 can be obtained, wherein the survival probability alignment chart corresponds to different risk score ranges of relative expression quantity, 1p19q state and pathology grade of the prognosis related genes. Calculating the accumulated sum of the risk scores corresponding to the 3 variables as a total risk score, and drawing a vertical line of a 1-year survival rate line of the patient at the position of the total risk score in the survival probability alignment chart, wherein the intersection point of the vertical line and the 1-year survival rate line of the patient is the probability that the survival period of the patient is 1 year; the probability of a patient survival of 2 years, 3 years or 5 years can be obtained using the same method. Different total risk scores correspond to probabilities of different patient survival being 1 year, 2 years, 3 years, and 5 years.
Example 2
This example was used to verify the predictive accuracy of the lower-grade glioma patient survival prediction system of example 1
A set of transcriptome expression profile data from 171 lower-grade glioma patients from china was collected as an internal validation set; a set of 415 lower-grade glioma patients from the united states transcriptome expression profile data served as the external validation set.
Survival predictions were made using the lower-grade glioma patient survival prediction system provided in example 1, and the predictions were compared for consistency with the actual survival of the patient.
The results show that the lower-grade glioma patient survival prediction system of example 1 had a consistency of 0.79 for the survival prediction results of the patients in the internal validation group with the actual survival time of the patients; the lower-grade glioma patient survival prediction system of example 1 had a consistency of 0.79 for the survival predictions of the patients in the external validation group with the actual survival of the patients. It can be seen that the survival time of the patient predicted by the survival time prediction system for the lower-grade glioma patient provided in example 1 has higher consistency with the actual survival time of the patient, and it can be seen that the survival time prediction system for the lower-grade glioma patient provided in example 1 has higher accuracy of the prediction result of the survival time of the lower-grade glioma patient.
Example 3
The accuracy of the predicted result of the survival prediction system for the lower-grade glioma patients provided in example 1 on the survival of 36 lower-grade glioma patients was verified by using PCR technology
Detecting the relative expression values of the 10 immune genes such as CRH-IFNB1, HOXA9-PRG3, IL10-IL9, IL9-PTH2, IL9-RETNLB, NKX2-5-PRLH, NKX3-2-UCN3, NR2C1-PTX3, PRLHR-REG1A and PRLHR-TRIM31 and the like on the constitutive genes by a quantitative PCR technology; patient pathology levels and 1p19q status were obtained by the pathology department and the patient survival probability was calculated using the nomogram model as control data.
Survival predictions were made using the lower grade glioma patient survival prediction system provided in example 1 and the predictions were compared for consistency with control data.
The results show that the survival prediction system of the lower-grade glioma patients in example 1 has a consistency of 0.75 with the control data for the survival prediction results of the patients in 36, further illustrating the higher accuracy of the survival prediction results of the lower-grade glioma patients provided in example 1.
In summary, the tumor patient survival prediction system provided by the disclosure predicts the survival of a tumor patient based on the relative expression level of the tumor prognosis related gene pair of the individual tumor patient, the variable is easy to obtain clinically, and the prediction result of the survival of the tumor patient can be obtained rapidly and accurately by using the variable, so that the time and energy of a user are saved, and the prediction accuracy can be further improved particularly after other clinical related indexes are matched.
The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details of the above embodiments, for example, the tumor may be any other tumor except for the lower grade glioma, the tumor prognosis related gene may be other kinds of genes except for the face and the reason, and various simple modifications may be made to the technical solution of the present disclosure within the scope of the technical concept of the present disclosure, and all the simple modifications are within the scope of the protection of the present disclosure.
In addition, the specific features described in the foregoing embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, the present disclosure does not further describe various possible combinations.
Moreover, any combination between the various embodiments of the present disclosure is possible as long as it does not depart from the spirit of the present disclosure, which should also be construed as the disclosure of the present disclosure.
Claims (7)
1. A tumor patient survival prediction system, comprising a computing device, an input device for inputting a risk score of a tumor prognosis related gene pair relative expression amount of an individual tumor patient, and an output device for outputting a survival probability of a tumor patient; the tumor patient comprises a glioma patient; the glioma is grade II glioma and/or grade III glioma;
wherein the tumor prognosis related gene pair is a gene pair with the relative size of the expression quantity of two genes in the gene pair and the survival time of a tumor patient; the gene pairs include CRH-IFNB1, HOXA9-PRG3, IL10-IL9, IL9-PTH2, IL9-RETNLB, NKX2-5-PRLH, NKX3-2-UCN3, NR2C1-PTX3, PRLHR-REG1A and PRLHR-TRIM31;
the computing device includes a memory and a processor; the memory stores a computer program to realize a modeling algorithm and an algorithm of a discriminant function shown in the formula (1); the modeling algorithm is a least partial square algorithm;
F（x，y）=f（x 1 ) F (y) formula (1),
wherein x is 1 A risk score representing the relative expression level of each pair of the tumor prognosis-related genes, y representing the survival of the tumor patient, f (x 1 ) A total risk score representing the relative expression level of the tumor prognosis-related gene, F (y) representing a survival rate curve function with a survival period of y years, and F (x, y) representing a probability of the tumor patient with a survival period of y years;
the input device is also used for inputting 1p19q state risk scores and pathology level risk scores of the tumor patient individuals;
the computing means is also for implementing an algorithm of a discriminant function as shown in equation (2),
F（x，y）=[f（x 1 ）+ f（x 2 ）+ f（x 3 ）]f (y) formula (2),
wherein x is 1 A risk score, x, representing the relative expression level of each pair of said tumor prognosis-related genes 2 Representing the 1p19q state risk score, x 3 Representing the pathological grade risk score, y representing the survival of the tumor patient, f (x) 1 ）+ f（x 2 ）+ f（x 3 ) Representing the total risk score, F (y) representing a survival curve function with a survival of y years, F (x, y) representing the probability of the tumor patient with a survival of y years;
f（x 2 ）=bx 2 wherein b represents a 1p19q state risk score coefficient;
f（x 3 ）=cx 3 wherein c represents a pathology level risk score coefficient;
relative expression level risk score coefficient a of gene pair CRH-IFNB1 1 Risk score coefficient a for relative expression level of HOXA9-PRG3 of-0.00332 gene 2 Risk score coefficient a for relative expression level of 0.124745 and gene to IL10-IL9 3 Risk score coefficient a for relative expression level of 0.205056 gene to IL9-PTH2 4 Risk score coefficient a for relative expression level of-0.11688 gene to IL9-RETNLB 5 Risk score coefficient a of relative expression level of-0.30775 gene to NKX2-5-PRLH 6 Risk score coefficient a for relative expression level of 0.035793 gene to NKX3-2-UCN3 7 Risk score coefficient a for relative expression level of 0.086457, gene pair NR2C1-PTX3 8 Risk score coefficient a for relative expression level of-0.49552, gene pair PRLHR-REG1A 9 Risk score coefficient a of relative expression amount of-0.24355 and gene pair PRLHR-TRIM31 10 Is-0.33529;
the 1p19q state risk score coefficient b is-1.1591;
the risk score coefficient c of the pathology level is 1.0261.
2. The tumor patient survival prediction system according to claim 1, wherein the calculation method of the risk score of the tumor prognosis-related gene to the relative expression amount is as follows:
in any of the tumor prognosis-related gene pairs, if the expression level of the first gene is greater than the expression level of the second gene, the relative expression level risk score x of the gene pair 1i Is the first relative expression levelRisk score; if the expression level of the first gene is less than or equal to the expression level of the second gene, the relative expression level risk score x of the gene pair 1i Risk score for the second relative expression quantity; wherein x is 1i Refers to the risk score of the i-th relative expression level of the tumor prognosis related gene pair.
3. The tumor patient survival prediction system of claim 2, wherein the first relative expression risk score has a value greater than 0 and the second relative expression risk score has a value less than or equal to 0.
4. The tumor patient survival prediction system of claim 1, wherein the method of calculating the 1p19q state risk score comprises:
when the 1p19q chromosome is in a deletion state, the 1p19q state risk score x 2 A first 1p19q state risk score; when the 1p19q chromosome is in a complete state, the 1p19q state risk score x 2 A second 1p19q state risk score;
the method for calculating the pathological level risk score comprises the following steps:
at a pathology level of WHO class II, the pathology level risk score x 3 A first pathology level risk score; when the pathology level is WHO III level, the pathology level risk score x 3 And a second pathology level risk score.
5. The tumor patient survival prediction system of claim 4, wherein the first 1p19q status risk score has a value greater than 0 and the second 1p19q status risk score has a value less than or equal to 0;
the value of the first pathological level risk score is smaller than or equal to 0, and the value of the second pathological level risk score is larger than 0.
6. The tumor patient survival prediction system according to claim 1, wherein the computing device is capable of creating a survival probability alignment chart and calculating a total risk score based on a risk score of the tumor prognosis related gene input by the input device for a relative expression amount, the 1p19q state risk score, and the pathology level risk score, the total risk score being a cumulative sum of the risk score of the tumor prognosis related gene input by the input device for a relative expression amount, the 1p19q state risk score, and the pathology level risk score, and calculating a tumor patient survival prediction value from the total risk score;
the method for establishing the survival probability alignment chart based on the risk scores of the tumor prognosis related genes, the 1p19q state risk scores and the pathology level risk scores input by the input device is to apply an R language RMS operation package to complete alignment chart visualization of a Cox regression model.
7. The tumor patient survival prediction system according to any one of claims 1-6, further comprising a detection device for the relative expression level of the tumor prognosis related gene pair, wherein the detection device comprises a detection chip and a chip signal reader for the relative expression level of the tumor prognosis related gene pair.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911222137.0A CN110993104B (en) | 2019-12-03 | 2019-12-03 | Tumor patient lifetime prediction system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911222137.0A CN110993104B (en) | 2019-12-03 | 2019-12-03 | Tumor patient lifetime prediction system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110993104A CN110993104A (en) | 2020-04-10 |
| CN110993104B true CN110993104B (en) | 2023-06-30 |
Family
ID=70089656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911222137.0A Active CN110993104B (en) | 2019-12-03 | 2019-12-03 | Tumor patient lifetime prediction system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110993104B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111863159B (en) * | 2020-06-01 | 2022-02-25 | 中山大学孙逸仙纪念医院 | Establishment method of line chart model for predicting curative effect of tumor immunotherapy |
| CN112133369B (en) * | 2020-08-26 | 2023-09-22 | 吴安华 | System for evaluating prognosis of tumor patient based on active oxygen and drug sensitivity evaluation and improvement method |
| CN112349421A (en) * | 2020-10-26 | 2021-02-09 | 中国医学科学院肿瘤医院 | Treatment prognosis evaluation model for lung cancer patient |
| CN112635063B (en) * | 2020-12-30 | 2022-05-24 | 华南理工大学 | Comprehensive lung cancer prognosis prediction model, construction method and device |
| CN113470782A (en) * | 2021-07-05 | 2021-10-01 | 复旦大学附属中山医院 | Model for predicting clinical prognosis before initial unresectable hepatocellular carcinoma surgery |
| CN114990222B (en) * | 2022-07-06 | 2023-07-18 | 山东大学第二医院 | Low-grade glioma patient total survival prediction model |
| CN117373534B (en) * | 2023-10-17 | 2024-04-30 | 中山大学孙逸仙纪念医院 | Triple negative breast cancer prognosis risk assessment system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101173313A (en) * | 2006-09-19 | 2008-05-07 | 天津医科大学附属肿瘤医院 | Mammary cancer diversion and prognosis molecule parting gene group, gene chip producing and using method |
| CN106874647A (en) * | 2017-01-06 | 2017-06-20 | 吴安华 | A kind of Gliblastoma patient survival forecasting system |
| CN108389624A (en) * | 2018-01-31 | 2018-08-10 | 浙江大学 | Duodenal adenocarcinoma prognosis prediction system based on COX proportional hazard models |
| CN108424969A (en) * | 2018-06-06 | 2018-08-21 | 深圳市颐康生物科技有限公司 | A kind of biomarker, the method for diagnosing or estimating mortality risk |
| CN108648826A (en) * | 2018-05-09 | 2018-10-12 | 中国科学院昆明动物研究所 | A kind of cancer of pancreas personalization prognostic evaluation methods based on multi-gene expression characteristic spectrum |
| CN108922628A (en) * | 2018-04-23 | 2018-11-30 | 华北电力大学 | A kind of Prognosis in Breast Cancer survival rate prediction technique based on dynamic Cox model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080275652A1 (en) * | 2005-05-13 | 2008-11-06 | Universite Libre De Bruxelles | Gene-based algorithmic cancer prognosis |
-
2019
- 2019-12-03 CN CN201911222137.0A patent/CN110993104B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101173313A (en) * | 2006-09-19 | 2008-05-07 | 天津医科大学附属肿瘤医院 | Mammary cancer diversion and prognosis molecule parting gene group, gene chip producing and using method |
| CN106874647A (en) * | 2017-01-06 | 2017-06-20 | 吴安华 | A kind of Gliblastoma patient survival forecasting system |
| CN108389624A (en) * | 2018-01-31 | 2018-08-10 | 浙江大学 | Duodenal adenocarcinoma prognosis prediction system based on COX proportional hazard models |
| CN108922628A (en) * | 2018-04-23 | 2018-11-30 | 华北电力大学 | A kind of Prognosis in Breast Cancer survival rate prediction technique based on dynamic Cox model |
| CN108648826A (en) * | 2018-05-09 | 2018-10-12 | 中国科学院昆明动物研究所 | A kind of cancer of pancreas personalization prognostic evaluation methods based on multi-gene expression characteristic spectrum |
| CN108424969A (en) * | 2018-06-06 | 2018-08-21 | 深圳市颐康生物科技有限公司 | A kind of biomarker, the method for diagnosing or estimating mortality risk |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110993104A (en) | 2020-04-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110993104B (en) | Tumor patient lifetime prediction system | |
| Long et al. | A four‐gene‐based prognostic model predicts overall survival in patients with hepatocellular carcinoma | |
| Chen et al. | Pan-cancer integrative histology-genomic analysis via multimodal deep learning | |
| Van Calster et al. | Extending the c‐statistic to nominal polytomous outcomes: the Polytomous Discrimination Index | |
| CN105492630A (en) | Assessment of the pi3k cellular signaling pathway activity using mathematical modelling of target gene expression. | |
| Velmahos et al. | Using deep learning to identify bladder cancers with FGFR‐activating mutations from histology images | |
| Li et al. | Identification of novel long non-coding RNA biomarkers for prognosis prediction of papillary thyroid cancer | |
| CN112735592B (en) | Construction method and application method of lung cancer prognosis model and electronic equipment | |
| CN115699204A (en) | Clinical predictor based on multiple machine learning models | |
| Wang et al. | Denoising autoencoder, a deep learning algorithm, aids the identification of a novel molecular signature of lung adenocarcinoma | |
| CN111440869A (en) | DNA methylation marker for predicting primary breast cancer occurrence risk and screening method and application thereof | |
| Ma et al. | Expression profiles analysis of long non-coding RNAs identified novel lncRNA biomarkers with predictive value in outcome of cutaneous melanoma | |
| CN110603597A (en) | System and method for biomarker identification | |
| Lv et al. | TransSurv: transformer-based survival analysis model integrating histopathological images and genomic data for colorectal cancer | |
| Phan et al. | omniBiomarker: a web-based application for knowledge-driven biomarker identification | |
| Niu et al. | Predicting tumor mutational burden from lung adenocarcinoma histopathological images using deep learning | |
| CN111161884A (en) | Disease prediction method, device, equipment and medium for unbalanced data | |
| Wu et al. | DeepRetention: a deep learning approach for intron retention detection | |
| CN116525114A (en) | Renal clear cell carcinoma prognosis prediction model based on PDK1 combined immunity | |
| Yang et al. | Automated facial recognition for Noonan syndrome using novel deep convolutional neural network with additive angular margin loss | |
| CN111383766A (en) | Computer data processing method, device, medium and electronic equipment | |
| KR102371655B1 (en) | Device, Method of Calculating Prostate Cancer Genetic Risk Score Based on Individual Weights for each Genetic Variation and Recording Medium thereof | |
| CN114672569A (en) | Tryptophan metabolism gene-based liver cancer prognosis evaluation method | |
| KR102102848B1 (en) | Prostate cancer risk score calculator, and method of the above calculator | |
| Han et al. | Molecular bases of morphometric composition in Glioblastoma multiforme |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |