CN115982644A

CN115982644A - Esophageal squamous cell carcinoma classification model construction and data processing method

Info

Publication number: CN115982644A
Application number: CN202310063027.4A
Authority: CN
Inventors: 刘芝华; 陈洪岩
Original assignee: Cancer Hospital and Institute of CAMS and PUMC
Current assignee: Cancer Hospital and Institute of CAMS and PUMC
Priority date: 2023-01-19
Filing date: 2023-01-19
Publication date: 2023-04-18
Anticipated expiration: 2043-01-19
Also published as: CN115982644B

Abstract

The invention discloses a method, a system, equipment and a computer readable storage medium for constructing an esophageal squamous cell carcinoma classification model and processing data, wherein the method comprises the following steps: obtaining sequencing data of a training set sample and life cycle conditions corresponding to the sample; extracting a DDR channel gene set and a gene expression condition thereof from sequencing data of the training set sample; carrying out selective processing on the DDR path gene set to obtain a path related to survival rate and a gene expression condition of the path related to survival rate; the survival-related pathway comprises one or more of the following: MMR access, NER access, FA access and NHEJ access; and carrying out cluster analysis on the training set samples based on the life cycle condition to obtain different classification subtypes, representing the path of each group of classification subtypes and the gene expression condition thereof, and obtaining a classification model.

Description

Esophageal squamous cell carcinoma classification model construction and data processing method

Technical Field

The invention relates to the field of data analysis, in particular to a method and a system for constructing and processing an esophageal squamous cell carcinoma classification model.

Background

Esophageal Squamous Cell Carcinoma (ESCC) is a malignant tumor that threatens human health. The five-year survival rate of ESCC patients is less than 20% in developed countries and less than 5% in many developing countries. Notably, some patients with primary esophageal cancer often relapse rapidly after esophageal resection, and the prognosis of these patients remains poor. To date, no accurate molecular biomarkers can predict the development of these primary ESCC patients, leading to inadequate clinical management. Therefore, there is an urgent need to determine new prognostic biomarkers for primary ESCC.

Multiple synergistic repair mechanisms can rapidly and properly repair DNA damage in normal cells; DNA double strand breaks are repaired mainly by Homologous Recombination (HR) and non-homologous end joining (NHEJ), and DNA single strand breaks are repaired mainly by mismatch repair (MMR) and nucleotide excision repair pathway (NER). DNA Damage Repair (DDR) defects can lead to the accumulation of DNA damage and genomic instability, the generation of neoantigens, and the upregulation of immune checkpoints, ultimately altering immune balance in the Tumor Microenvironment (TME). Interestingly, DDR deficiency becomes an important determinant of anti-tumor immune response by affecting antigenicity, adjuvanticity, and reactivity, which may contribute to the response of immunotherapy. Recent studies have revealed the potential of some DDR-based biomarkers in predicting immunotherapeutic responses; however, the value of DDR-related features for prognostic evaluation and personalized immunotherapy has not yet been fully elucidated. Therefore, it is crucial to reveal a correlation between changes in the tumor DDR pathway and prognosis.

Disclosure of Invention

The present invention is directed to solving at least one of the problems of the prior art. The invention provides a construction method of an esophageal squamous cell carcinoma classification model, which comprises the steps of screening out a DDR (double data rate) channel gene set and a gene expression condition thereof by using sequencing data of a sample, carrying out cluster analysis on the sample according to the life cycle condition of the sample to obtain a DDR-active subtype and a DDR-silent subtype, representing the DDR channel gene set and the gene expression condition of 2 subtypes and obtaining the classification model; the method provided by the invention is used for typing and prognosis evaluation of the primary ESCC by processing and analyzing the related data based on the classification model, and deeply mining the life law hidden behind the biological data to solve the related life science problem.

The first aspect of the application discloses a method for constructing an esophageal squamous cell carcinoma classification model, which comprises the following steps:

obtaining sequencing data of a training set sample and life cycle conditions corresponding to the sample;

extracting a DDR channel gene set and a gene expression condition thereof from sequencing data of the training set sample;

carrying out selective processing on the DDR path gene set to obtain a path related to survival rate and a gene expression condition of the path related to survival rate; the survival-related pathway comprises one or more of the following: MMR access, NER access, FA access and NHEJ access;

and carrying out cluster analysis on the training set samples based on the life cycle condition to obtain different classification subtypes, representing the path of each group of classification subtypes and the gene expression condition thereof, and obtaining a classification model.

The DDR pathway gene set includes: a BER path, an MMR path, an NER path, an FA path, an HR path and an NHEJ path;

the method for cluster analysis comprises the following steps: a consistency clustering algorithm;

optionally, the method for selecting processing includes: univariate Cox regression analysis;

optionally, the sequencing data of the training set sample comprises: RNA-seq data for primary ESCC tumor tissue samples and metastatic ESCC tumor tissue samples.

The construction method further comprises the following steps: based on the gene expression condition of the survival rate related pathway, obtaining a DDR gene set related to a survival result and a corresponding gene expression condition by using a univariate regression analysis method; processing the DDR gene set related to the survival result and the corresponding gene expression condition by using a multivariate analysis method to obtain a prognosis prediction gene and the gene expression condition of the prognosis prediction gene;

and carrying out cluster analysis on the training set samples based on the life cycle condition to obtain different classification subtypes, prognosis prediction genes representing each group of classification subtypes and gene expression conditions thereof, and obtaining a classification model.

The different classification subtypes of the classification model include: DDR-active subtype and DDR-silent subtype; the DDR-active subtype corresponds to the gene expression condition of a path with high survival rate, and the DDR-silent subtype corresponds to the gene expression condition of a path with low survival rate;

optionally, the prognostic prediction genes include: BRCA1 gene and HFM1 gene; the DDR-active subtype has high expression level corresponding to BRCA1 gene, and the DDR-silent subtype has high expression level corresponding to HFM1 gene.

In a second aspect, the present application discloses a method for processing esophageal squamous cell carcinoma data, comprising:

obtaining sequencing data of a sample to be detected;

inputting the sequencing data of the sample to be tested into the classification model disclosed by the first aspect of the application to obtain the classification results of the DDR-active subtype and the DDR-silent subtype;

optionally, the method further includes: predicting the survival rate of the sample to be tested based on the classification result; outputting a result with high survival rate of the sample to be detected based on the DDR-active subtype classification result; and outputting a result of low survival rate of the sample to be detected based on the classification result of the DDR-silent subtype.

In a third aspect of the present application, a method for processing esophageal squamous cell carcinoma data is disclosed, which comprises:

acquiring gene expression data of a sample to be detected; the gene expression data of the sample to be detected comprises gene expression data of one or more of the following genes: BRCA1 gene, HFM1 gene;

inputting the gene expression data of the sample to be detected into the classification model disclosed in the first aspect of the application to obtain a classification result;

a fourth aspect of the present application discloses a system for processing esophageal squamous cell carcinoma data, comprising:

the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring sequencing data of a sample to be detected;

and the output unit is used for inputting the sequencing data of the sample to be detected into the classification model disclosed in the first aspect of the application to obtain the classification results of the DDR-active subtype and the DDR-silent subtype.

A fifth aspect of the present application discloses an apparatus for processing esophageal squamous cell carcinoma data, the apparatus comprising: a memory and a processor;

the memory is to store program instructions; the processor is used for calling program instructions and executing the method for processing the esophageal squamous cell carcinoma data disclosed in the second aspect and/or the third aspect of the application when the program instructions are executed.

A sixth aspect of the present application discloses a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method for processing esophageal squamous cell carcinoma data disclosed in the second and/or third aspects of the present application.

The application has the following beneficial effects:

1. the application innovatively discloses a model construction method for typing primary ESCC according to a DDR path gene set and a gene expression condition thereof, and a classification model of 2 classification results of a DDR-active subtype and a DDR-silent subtype is obtained; meanwhile, in the process of model construction, two independent prognostic biomarkers BRCA1 and HFM1 are also determined, the classification model can be used for effectively predicting the subsequent survival rate of the primary ESCC patient with frequent and rapid relapse and poor prognosis, new clues and new perspectives are provided for the identification of the novel DDR-based molecular subtype as tumor heterogeneity, and the potential clinical significance of the treatment and management strategy of the primary ESCC patient with the DDR-silent subtype is revealed.

2. The method is used for carrying out clinical prognosis evaluation on the patient, life rules hidden behind biological data are mined from a deep level, and the accuracy and the depth of data analysis are greatly improved from a plurality of dimensions such as gene information and channel information of a biological population.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic flow chart diagram of a method for processing esophageal squamous cell carcinoma data provided by a second aspect of an embodiment of the present invention;

FIG. 2 is a schematic diagram of an esophageal squamous cell carcinoma data processing and analyzing device provided by the embodiment of the invention;

FIG. 3 is a schematic flow chart of a system for processing and analyzing esophageal squamous cell carcinoma data provided by an embodiment of the invention;

FIG. 4 is a diagram of cluster analysis of ESCC tumors based on DDR gene profiling provided by embodiments of the present invention;

FIG. 5 is a graph showing the results of BRCA1 and HFM1 modulating DNA damage response in ESCC as provided in the examples of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.

In some of the flows described in the present specification and claims and in the above figures, a number of operations are included that occur in a particular order, but it should be clearly understood that these operations may be performed out of order or in parallel as they occur herein, with the order of the operations being indicated as 101, 102, etc. merely to distinguish between the various operations, and the order of the operations by themselves does not represent any order of performance. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor limit the types of "first" and "second" to be different.

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a schematic flow chart of a processing method of esophageal squamous cell carcinoma data provided by a second aspect of the embodiment of the invention, specifically, the method comprises the following steps:

101: obtaining sequencing data of a sample to be detected;

in one embodiment, the sequencing data of the test sample is RNA-seq data of a primary ESCC patient. Primary is relative to secondary and metastatic. That is, a disease occurs first in a tissue or organ for which the disease is primary. To give an example: primary hepatocellular carcinoma, the primary one, is the cancer of the liver cells, while secondary liver cancer, the cancer of other parts, which is transferred to the liver by the blood flow or lymphatic route, is in other tissues or organs than the liver. Primary ESCC patients in this example refer to patients who have undergone surgical resection of a primary tumor, followed by radiation therapy, with or without chemotherapy.

In one embodiment, RNA-seq or transcriptome sequencing techniques are used to perform sequencing analysis using high throughput sequencing techniques to reflect the expression levels of mRNA, smallRNA, noncodingRNA, etc., or some of them. In the past decade, RNA-Seq technology has evolved rapidly and has become an indispensable tool for analyzing differential gene expression/mRNA variable splicing at the transcriptome level. With the development of the next generation sequencing technology, the application range of the RNA-Seq technology becomes wider: in the field of RNA biology, RNA-Seq can be applied to single cell gene expression/protein expression/RNA structure analysis; secondly, the concept of spatial transcriptome is also gradually emerging. The long-read long/direct RNA-Seq technique and better data analysis computational tools help biologists to use RNA-Seq to deepen understanding of RNA biology-e.g., when and where transcription begins; how the folding and intermolecular action in vivo affect the RNA function, and the like.

A transcriptome is the collection of all transcripts produced by a certain species or specific cell type. Transcriptome research can study gene functions and gene structures from the whole level, reveal molecular mechanisms in specific biological processes and disease occurrence processes, and has been widely applied in the fields of basic research, clinical diagnosis, drug research and development, and the like.

In one embodiment, the test sample is a primary ESCC patient clinically used to receive a prognostic assessment.

102: inputting the sequencing data of the sample to be detected into the constructed classification model to obtain the classification results of DDR-active subtype and DDR-silent subtype;

in one embodiment, the method further comprises: predicting the survival rate of the sample to be tested based on the classification result; outputting a result with high survival rate of the sample to be detected based on the classification result of the DDR-active subtype; outputting a result of low survival rate of the sample to be detected based on the classification result of the DDR-silent subtype;

in one embodiment, the method for constructing the classification model comprises the following steps:

extracting a DDR channel gene set and a gene expression condition thereof from sequencing data of the training set sample; the training set samples included RNA-seq data for tumor tissues of 82 primary ESCCs and 73 ESCCs with lymph node metastasis; the patients received surgical resection and lymph node dissection of the primary tumor, followed by radiation therapy, with or without chemotherapy. The data for the 155 patients were from the ESCC cohort of the tumor Hospital (SCH) of Shanxi province, the RNA-seq data for the SCH cohort were deposited in Gene Expression Omnibus (GEO) under the accession number GSE53625, and clinical and pathological data were determined for 97 patients by retrospective examination of SCH electronic medical records, ending the follow-up period in 2019/06 months. RNA-seq data analysis for the HiSeq Illumina platform was collected from UCSC Xena atlas (https:// Xena browser. Net/datapages /), RNA-seq data covering TPM levels and log2 (x + 1) normalization;

and performing cluster analysis on the training set samples based on the life cycle condition to obtain different classification subtypes, representing the path of each group of classification subtypes and the gene expression condition thereof, and obtaining a classification model.

In one example, to characterize DDR subtypes, differential Expression (DE) analysis was first performed on DDR subtypes using R package limma (v3.50.3) to determine subtype-specific genes. Differentially Expressed Genes (DEG) were defined as log fold change (logFC) < = -1 or > =1 and adjusted P value <0.05. Then, a pathway enrichment analysis was performed on the DEG's from a select set of marker pathways from MSigDB (GenBank database: https:// www. Jianshu. Com/p/99369b2f7a7 d) to identify enriched pathways in DDR subtypes, as performed by the R-packet Cluster analysis program Cluster profiler (version 4.2.2).

In one embodiment, the gene expression profile of the survival-related pathway comprises gene expression profiles of one or more of the following genes: POLD1, POLD2, POLD3, POLD4, MSH2, MSH3, MSH6, MLH1, MLH3, PMS1, PMS2, MSH4, MSH5, EXO1, HMGB1, LIG1, PCNA, RFC2, RFC4, RFC3, RFC5, RFC1, RPA2, RPA3, RPA4, POLD1, POLD2, POLD3, POLD4, PCNA, RFC1, RFC2, RFC3, RFC4, RFC5, POLE2, POLE3, POLE4, POLK, CUL4A, DDB1, DDB2 RBX1, CUL4A, DDB1, DDB2, RBX1, CETN2, RAD23B, XPC, POLR2A, POLR2B, POLR2C, POLR2D, POLR2E, POLR2F, POLR2G, POLR2H, POLR2I, POLR2J, POLR2K, POLR2L, CUL3, CUL5, ERCC1, ERCC4, ERCC5, LIG1, TCEB2, TCEB3, UVSSA, XPA, RPA1, RPA2, RPA3, RPA4, CDK7, ERCC2, ERCC3, GTF2H1, GTF2H2 GTF2H3, GTF2H4, GTF2H5, MNAT1, ERCC6, ERCC8, LIG3, RAD23A, XAB2, XRCC1, GADD45A, GADD45G, BLM, RMI2, TOP3A, TOP3B, BARD1, BRCA2, BRIP1, PALB2, FAAP100, FAAP24, FANCA, FANCB, FANCC, FANCE, FANCF, FANCG, FANCL, FANCM, APITD1, HES1, STRA13, UBE2T, FANCD2, FANCI, BRE, CCDC98, DNA2, FAN1, HELQ KAT5, RAD51C, TELO2, USP1, WDR48, APLF, ATM, MDC1, MRE11A, NBN, PARP3, RAD50, RNF168, RNF8, TP53BP1, DCLRE1C, LIG4, NHEJ1, PRKDC, XRCC5, XRCC6, LIG4, NHEJ1, XRCC4, PRKDC, XRCC5, XRCC6, PNKP, POLL, MRE11A, RAD50, DNTT, POLL, APLF, APTX, DCLRE1C, PARG, XRCC2, XRCC3;

optionally, the DDR pathway gene set comprises: BER pathway (base excision repair, n = 43), MMR pathway (mismatch repair, n = 27), NER pathway (nucleotide excision repair, n = 70), FA pathway (fanconi anemia, n = 36), HR pathway (homologous recombination, n = 55), and NHEJ pathway (non-homologous end joining, n = 37);

in one embodiment, the method of cluster analysis is: a consistency clustering algorithm; consistent clustering is also known as consensus clustering, which is a method of aggregating the results of multiple clustering algorithms, also known as cluster integration or clustering. It is meant that many different (input) clusters have been obtained for a particular data set and that it is desirable to find a single (consistent) cluster, in a sense more appropriate than existing clusters. Consistent clustering is therefore a problem of reconciling clustering information about the same data set from different sources or different runs of the same algorithm. This clustering procedure was performed using R-pack consensus, 1000 iterations and 90% resampling. The core algorithm is a k-means algorithm based on Euclidean distance, and a single algorithm cannot be realized.

optionally, the sequencing data of the training set sample comprises: RNA-seq data for primary ESCC tumor tissue samples and metastatic ESCC tumor tissue samples. By analyzing RNA-seq data for primary and metastatic ESCC tumor tissue samples, DDR pathway analysis established that DDR active and DDR silent subtypes have independent prognostic value in primary ESCCs, but not in metastatic ESCCs.

In one embodiment, the construction method further comprises: based on the gene expression condition of the survival rate related pathway, obtaining (8) DDR gene sets related to the survival result and corresponding gene expression conditions by using a univariate regression analysis method; processing the DDR gene set related to the survival result and the corresponding gene expression condition by using a multivariate analysis method to obtain a prognosis prediction gene and a gene expression condition of the prognosis prediction gene; the sex, grade, smoking history and drinking history are controlled in the multivariate analysis;

and performing cluster analysis on the training set samples based on the life cycle condition to obtain different classification subtypes, representing prognosis prediction genes of each group of classification subtypes and gene expression conditions thereof, and obtaining a classification model.

The different classification subtypes of the classification model include: DDR-active subtype and DDR-silent subtype; the DDR-active subtype corresponds to the gene expression condition of a path with high survival rate, and the DDR-silent subtype corresponds to the gene expression condition of a path with low survival rate; no specific threshold value exists in the survival rate, and the survival rate is concluded through statistical comparison analysis between DDR-active subtype and DDR-silent subtype.

In one example, the correlation between DDR subtypes and ESCC survival with and without LNM (lymph node metastasis) was studied using a hierarchical analysis approach, and the results showed that the survival rate of primary ESCC tumors of DDRslient subtypes was the worst (log-rankp = 0.032) compared to primary and metastatic ESCC tumors of DDR-slient subtypes, but no significant difference was observed in the survival rate of metastatic ESCC tumors between DDR subtypes (log-rankp = 0.34). DDR pathway analysis established that DDR active subtypes and DDR silent subtypes have independent prognostic value in primary ESCCs, but not in metastatic ESCCs.

In one embodiment, to further validate the association between DDR subtype typing and survival outcome, we also summarized DDR subtypes for 74 tumors in the TCGA-ESCC cohort and 117 tumors in the Chen cohort. Consistent with the findings in this cohort, the DDR subtype-assisted survival prediction was only used for primary ESCC tumors, allowing identification of patient subgroups with good or poor outcome (TCGA-ESCC cohort, HR =0.075, 95-cent ci 0.008-0.674, log-rankp =0.004; HR =0.430, 95-ci 0.186-0.995, log-rankp =0.042 for Chen cohort), and no stratification of survival of ESCC tumors with LNM was possible. Multivariate Cox regression analysis indicates that the DDR subtype is a powerful predictor of survival outcome, and that it is independent of clinical variables, and emphasizes the value of the DDR subtype and its robustness in predicting survival outcomes for primary ESCC patients. The hierarchical analysis is to divide the population into different layers (sub-layers) according to certain characteristics, such as gender, age and the like, and analyze the relationship between exposure and diseases in each layer respectively. The purpose of the hierarchical analysis is to control confounding factors, adjust the interference of these factors — estimate the magnitude of the confounding factor's impact on the correlation between exposure factors and outcomes. The hierarchical analysis is to cope with the scenario of mean value failure. Wherein the TCGA-ESCC cohort included RNA-seq data for 74 patients, collected from UCSC Xena atlas (https:// Xena brown. Net/datapages /); the Chen cohort is the microarray data for 117 ESCC patients from the chinese academy of medical sciences and the beijing institute of cooperative medicine, with clinical data obtained from Gene Expression Omnibus (GEO, www.ncbi.nlm.nih.gov/GEO/query/acc.cgiac = GSE 53624).

Optionally, the prognostic predictive genes include: BRCA1 gene and HFM1 gene; the DDR-active subtype has high expression level corresponding to BRCA1 gene, and the DDR-silent subtype has high expression level corresponding to HFM1 gene; no specific threshold value exists for the level of gene expression, and the result is obtained by performing statistical comparison analysis between DDR-active subtype and DDR-silent subtype.

In one example, prognostic assessment of DDR genes in primary and metastatic ESCC tumor tissues using a meta assay with 3 independent cohorts, respectively, resulted in BRCA1 and HFM1 being predictors of survival outcome in primary ESCC patients, but not contributing to prognosis of metastatic ESCC; BRCA1 was identified as a favorable prognostic factor, with high expression associated with improved survival, with a combined HR of 0.22, while HFM1 is a risk factor, with increased expression associated with poor survival outcome, with a different aggregate HR of 4.41.

In one embodiment, in the presence of BRCA1, cells sense and repair DNA damage, maintain genomic integrity and prevent tumorigenesis. BRCA1 deficiency destroys normal DDR and leads to the accumulation of DNA damage. However, the role of HFM1, an ATP-dependent DNA helicase homolog, in DDR has not been studied. To determine the role of BRCA1 and HFM1 in ESCC cells DDR, a cellular model of cisplatin (DDP) and X-IR-induced DNA damage in vitro was constructed, using transient siRNA transfection to silence BRCA1 or HFM1 expression, and treating the cells with cisplatin (DDP) or X-IR. The knockout efficiency of BRCA1 and HFM1 is detected by Western blotting. For direct assessment of DDR, γ H2AX (a mature DNA DSB marker) was visualized by immunofluorescence. Spontaneous and DDP or IR induced γ H2AX foci were counted and analyzed. After DDP or X-IR treatment, γ H2AX accumulates. Furthermore, immunofluorescence analysis indicated a significant increase in endogenous γ H2AX accumulation in KYSE410 and KYSE450 cells following BRCA1 knockdown under IR and DDP treatment. In contrast, knock-out of HFM1 significantly reduced the number of γ H2AX foci in KYSE30 and KYSE450 cells treated with X-IR or DDP. These results indicate that deletion of BRCA1 results in DDR deficiency, supporting the role of BRCA1 as a favorable prognostic factor, while deletion of HFM1 promotes DDR, supporting the role of HFM1 as a prognostic risk factor.

optionally, the gene expression data of the test sample is the data of the primary ESCC patient.

Fig. 2 is a device for processing and analyzing esophageal squamous cell carcinoma data, which is provided by the embodiment of the invention and comprises: a memory and a processor; the memory is to store program instructions; the processor is configured to invoke program instructions, which when executed, are configured to perform a method of processing esophageal squamous cell carcinoma data as described above.

Fig. 3 is a system for processing and analyzing esophageal squamous cell carcinoma data, which comprises:

an obtaining unit 301, configured to obtain sequencing data of a sample to be detected;

the output unit 302 is configured to input the sequencing data of the sample to be tested into the classification model disclosed in the first aspect of the present application, so as to obtain classification results of DDR-active subtypes and DDR-silent subtypes.

The processing and analyzing system for esophageal squamous cell carcinoma data provided by the embodiment of the invention comprises:

an acquisition unit that acquires gene expression data of a sample to be tested; the gene expression data of the sample to be detected comprises gene expression data of one or more of the following genes: BRCA1 gene, HFM1 gene;

and the output unit is used for inputting the gene expression data of the sample to be detected into the classification model disclosed by the first aspect of the application to obtain a classification result.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out a method of processing esophageal squamous cell carcinoma data as described above.

FIG. 4 is a diagram of cluster analysis of ESCC tumor based on DDR gene mapping provided in the embodiment of the present invention, wherein,

(A) Heat map of DDR gene expression fold change between DDR subtypes. The red bar represents the DDR-active subtype and the green bar represents the DDR-silent subtype. The DDR subtypes are classified by consensus clustering. (B-D) Kaplan-Meier curves comparing OS (log rank test) for DDR-active subtype, DDR-silent subtype and transition subtype groups. HR and 95% CI were calculated by a two-sided Wald test using univariate Cox regression.

FIG. 5 is a graph showing the results of BRCA1 and HFM1 modulating DNA damage response in ESCC as provided in the examples of the present invention, wherein (A, B) KYSE410 and KYSE450 cells were transfected with BRCA1 siRNA, treated with 2. Mu.g/ml DDP, and analyzed by Western blotting for γ H2AX. (C, D) KYSE410 and KYSE450 cells were transfected with BRCA1 siRNA, exposed to IR (4 Gy), harvested at the indicated times, and analyzed for γ H2AX by Western blotting. Representative pictures and quantification of γ H2AX foci in (E, F) control and BRCA1 knockdown KYSE410 and KYSE450 cells, treated with 2 μ g/ml DDP for the indicated time. Data are representative of three independent experiments. Each point represents one cell and 50 cells per group were counted for this experiment using Image J. Error bars represent SD of this experiment. P values were determined by unpaired two-sided t-test. Representative pictures and quantification of γ H2AX foci in (G, H) control and BRCA1 knock-out KYSE410 and KYSE450 cells, treated with IR (4 Gy) for the indicated time. Data are representative of three independent experiments. Each point represents one cell and 50 cells per group were counted for this experiment using Image J. Error bars represent SD of this experiment. P values were determined by unpaired two-sided t-test. (I, J) KYSE30 and KYSE450 cells were transfected with HFM1 siRNA, treated with 2. Mu.g/ml DDP, and analyzed by Western blot

γ H2AX. (K, L) KYSE30 and KYSE450 cells were transfected with HFM1 siRNA, exposed to IR (4 Gy), harvested at the indicated times, and analyzed for γ H2AX by Western blotting. Representative pictures and quantification of γ H2AX foci in (M, N) control and HFM1 knockdown KYSE30 and KYSE450 cells, treated with 2 μ g/ml DDP for the indicated time. Data are representative of three independent experiments. Each point represents one cell and Image J counts 50 cells for each group of this experiment. Error bars represent SD of this experiment. P values were determined by unpaired two-sided t-test. Representative pictures and quantification of (O, P) control and HFM1 knockdown of KYSE30 and KYSE450 cells with IR (4 Gy) treatment of γ H2AX foci at the indicated times. Data are representative of three independent experiments. Each point represents one cell and Image J counts 50 cells for each group of this experiment. Error bars represent SD of this experiment. P values were determined by unpaired two-sided t-test.

The validation results of this validation example show that assigning an intrinsic weight to an indication can moderately improve the performance of the method relative to the default settings.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one type of logical functional division, and other divisions may be realized in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: read Only Memory (ROM), random Access Memory (RAM), magnetic or optical disks, and the like.

It will be understood by those skilled in the art that all or part of the steps in the method for implementing the above embodiments may be implemented by hardware that is instructed to implement by a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

While the invention has been described in detail with reference to specific embodiments thereof, it will be apparent to one skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims

1. A method for constructing an esophageal squamous cell carcinoma classification model comprises the following steps:

extracting a DDR channel gene set and a gene expression condition thereof from sequencing data of the training set sample; carrying out selective processing on the DDR path gene set to obtain a path related to survival rate and a gene expression condition of the path related to survival rate; the survival-related pathway comprises one or more of the following: MMR access, NER access, FA access and NHEJ access;

2. The method for constructing the esophageal squamous cell carcinoma classification model according to claim 1, wherein the DDR pathway gene set comprises one or more of the following genes: BER pathway, MMR pathway, NER pathway, FA pathway, HR pathway, and NHEJ pathway.

3. The method for constructing the esophageal squamous cell carcinoma classification model according to claim 1, wherein the method of cluster analysis is as follows: a consistency clustering algorithm;

4. The method of constructing an esophageal squamous cell carcinoma classification model according to claim 1, characterized in that said method of construction further comprises: based on the gene expression condition of the survival rate related pathway, obtaining a DDR gene set related to a survival result and a corresponding gene expression condition by using a univariate regression analysis method; processing the DDR gene set related to the survival result by using a multivariate analysis method to obtain a prognosis prediction gene and a gene expression condition of the prognosis prediction gene;

5. The method for constructing the classification model of esophageal squamous cell carcinoma according to any of claims 1-4, characterized in that the different classification subtypes of the classification model comprise: DDR-active subtype and DDR-silent subtype; the DDR-active subtype corresponds to the gene expression condition of a path with high survival rate, and the DDR-silent subtype corresponds to the gene expression condition of a path with low survival rate;

optionally, the prognostic predictive genes include: BRCA1 gene and HFM1 gene; the DDR-active subtype has high expression level corresponding to BRCA1 gene, and the DDR-silent subtype has high expression level corresponding to HFM1 gene.

6. A method of processing esophageal squamous cell carcinoma data, comprising:

obtaining sequencing data of a sample to be detected;

inputting the sequencing data of the sample to be tested into the classification model in claims 1-5 to obtain the classification results of DDR-active subtype and DDR-silent subtype;

optionally, the method further includes: predicting the survival rate of the sample to be tested based on the classification result;

outputting a result with high survival rate of the sample to be detected based on the classification result of the DDR-active subtype; and outputting a result of low survival rate of the sample to be detected based on the classification result of the DDR-silent subtype.

7. A method of processing esophageal squamous cell carcinoma data, comprising:

inputting the gene expression data of the sample to be detected into the classification model in claims 1-5 to obtain a classification result;

optionally, the gene expression data of the test sample is data of a primary ESCC patient.

8. A system for processing esophageal squamous cell carcinoma data, comprising:

the output unit is used for inputting the sequencing data of the sample to be tested into the classification model in the claims 1-5 to obtain the classification results of DDR-active subtype and DDR-silent subtype.

9. An apparatus for processing esophageal squamous cell carcinoma data, the apparatus comprising: a memory and a processor;

the memory is to store program instructions; the processor is configured to invoke program instructions for performing the method of processing esophageal squamous cell carcinoma data of claim 6 or 7 when the program instructions are executed.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of processing esophageal squamous cell carcinoma data of claim 6 or 7 as set forth above.