Construction and application of stomach cancer evolution visual diagnosis and treatment model driven by chemotherapy pressure

Abstract

The invention provides construction and application of a chemotherapy pressure-driven gastric cancer evolution visual diagnosis and treatment model. The construction comprises the following three steps: A, establishing a standardized sample library; B, constructing a tumor clone evolution network regulation and control table; and C, verifying and evaluating a key path. According to the method, dimensionality reduction and visual presentation are carried out on gastric cancer multi-dimensional omics data by applying various bioinformatics means, various graphic characteristics are associated with phenotypes such as clinical curative effect and outcome, a gastric cancer chemotherapy evolution characteristic spectrum is described in a panoramic manner, and a clone evolution rule of gastric cancer cells driven by chemotherapy pressure is explored. With'multi-point dynamic sampling 'and'multi-omics combination', and with a visual platform capable of establishing image mapping of the above characteristics and integrating radiomics, genome / transcriptome / proteomics, molecular pathology and clinical characterization as a target, directional clone evolution characteristics of gastric cancer driven by chemotherapy pressure are displayed in a panoramic manner; and decision support is provided for more reasonable clinical layout of a medication strategy.

Description

Technical Field

[0001] The present invention belongs to the technical field of tumor diagnosis and treatment models, and specifically relates to a method for constructing a visualized diagnosis and treatment model of gastric cancer evolution driven by chemotherapy pressure and its application. Background Art

[0002] At present, there is a relative lack of multidimensional cognition and clinical precision decision-making methods for advanced gastric cancer. Whether it is chemotherapy, molecular targeted therapy or immunotherapy, the clinical beneficiary population is relatively limited and long-term application eventually fails. The main reason is the heterogeneity of the tumor and the complex molecular evolution process driven by external treatment pressure. Complex heterogeneity is an important cause of treatment resistance. Gastric cancer is one of the most heterogeneous tumors in terms of both tissue morphology and molecular level. High heterogeneity is considered to be a characteristic of tumors overcoming evolutionary pressure, but it also means that heterogeneity itself is expected to be developed as a new therapeutic target; this also highlights the importance of developing quantitative technologies and models for evaluating tumor heterogeneity. If the dynamic evolution of tumor heterogeneity and the directed differentiation of drug resistance driven by chemotherapy drug pressure can be reflected through a visual interface, and pathological morphology, imaging, clinical characteristics and clinical outcomes can be integrated to fully display the significance of this interaction on clinical outcomes and treatment decisions, it will be better for future exploration of targeted intervention treatment methods for tumor clonal evolution and provide a comprehensive and detailed clinical data basis.

[0003] Tumors involve multiple abnormalities at the levels of DNA, RNA, proteins, and metabolites. Over the past few decades, single-omics research on tumors has achieved remarkable results, but the exact mechanism of tumor occurrence and development is still unclear. In order to more systematically reveal the process and mechanism of tumor occurrence and development, multi-omics research has emerged, promoting the transformation of the tumor research paradigm from a single-parameter model to a multi-parameter system model. However, faced with the complex data of various omics, we often have no idea where to start, let alone extract effective data information. This requires us to integrate the data obtained, including statistics, machine learning techniques, or analysis of regulatory or model models, in order to achieve a deeper understanding of cancer.

[0004] Therefore, under this clinical need, it is particularly urgent to construct a visual evolutionary model driven by chemotherapy pressure in gastric cancer. Summary of the invention

[0005] The present invention is aimed at the above-mentioned problems, and aims to use a variety of bioinformatics methods to reduce the dimension and visualize the multidimensional omics data of gastric cancer, establish associations between various graphic features and phenotypes such as clinical efficacy and prognosis, and comprehensively depict the evolutionary characteristic spectrum of gastric cancer chemotherapy, explore the clonal evolution rules of gastric cancer cells driven by chemotherapy pressure, and provide decision support for more reasonable clinical layout of drug use strategies.

[0006] At the same time, in view of the huge gap between the current clinical practice based on "single biopsy" and "single-omics" and the diversity and dynamic changes presented by tumor heterogeneity, the present invention uses "multi-point dynamic sampling" and "multi-omics combination" to establish a visualization platform that can mirror the above characteristics, integrating imaging omics, genomics / transcriptomics / proteomics, molecular pathology and clinical characterization, to comprehensively display the characteristics of directional clonal evolution of gastric cancer driven by chemotherapy pressure.

[0007] The present invention obtains markers or drug target protein markers (groups) that can indicate drug resistance, analyzes the evolution of tumors driven by chemotherapy drug pressure, and provides a visual diagnosis and treatment model for clinical implementation of real-time and accurate decision-making for targeted intervention in tumor evolution and full-process anti-tumor management.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0009] In a first aspect, the present invention provides a method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure, characterized in that it comprises the following steps:

[0010] A. Establish a standardized sample library

[0011] Dynamically collect peripheral blood from gastric cancer patients who initially use basic chemotherapy drugs before chemotherapy, at the time of best efficacy, and at disease progression, obtain circulating tumor cells (CTCs) from them, culture them, and amplify them, and perform plasma tests on the peripheral blood collected at these three time points; collect paired response-non-response, before chemotherapy-best efficacy-disease progression tumor tissues, separate cells, and culture and amplify them; and establish a clinical database based on basic patient information.

[0012] B. Constructing a tumor clone evolution network control platform

[0013] Through circulating omics, imaging omics, and genomic / transcriptomic / proteomic technologies, the heterogeneity of individuals with different therapeutic effects is analyzed, and a multi-dimensional omics chemotherapy evolution map is drawn; through bioinformatics analysis combined with cancer cell branch mutation and trunk mutation analysis, a tumor clone evolution network control station is constructed to find and fit the key regulatory pathways involved in chemotherapy resistance.

[0014] C. Verification and critical path assessment

[0015] Differentially expressed factors are screened based on the key regulatory pathways, and the clinical value of the key pathways in the evolutionary laws driven by chemotherapy drugs is evaluated after verification.

[0016] The preferred steps of the above three steps are as follows:

[0017] In step A, the basic chemotherapy drug is selected from platinum chemotherapy drugs, fluorouracil chemotherapy drugs or paclitaxel chemotherapy drugs;

[0018] Tumor cells obtained from peripheral blood and tumor tissues were cultured and expanded using two-dimensional and three-dimensional systems;

[0019] When plasma testing is performed on peripheral blood, free DNA, plasma exosomes, and cytokines are tested;

[0020] The patient's basic information includes treatment status, progression time, prognosis assessment results, pathological data, imaging images and data.

[0021] In step B, the method of mapping the evolution of chemotherapy by cyclic omics is as follows: (1) performing single-cell sequencing on the tumor cells captured in step (A), analyzing the differences in the gene spectra of tumor cells at three time points before chemotherapy, at the best therapeutic effect, and at the time of disease progression by bioinformatics, evaluating the direction of clonal evolution, and obtaining an evolutionary gene set; (2) isolating the plasma at the three time points, detecting cfDNA, plasma exosomes, and cytokines, and looking for differential expression changes in different individuals and at different therapeutic effects, mapping the tumor evolution, exploring the evolutionary laws, and mining potential tumor evolution pathways driven by chemotherapy drugs.

[0022] The method of mapping the evolution of chemotherapy through radiomics is as follows: using radiomics technology, extract the imaging data of patients paired with blood and tissue specimens before chemotherapy, at the best therapeutic effect and at different stages of tumor progression, and use machine learning to analyze the imaging differences between individuals with different therapeutic effects; at the same time, combined with circulating omics data, bioinformatics technology is used to analyze the correlation between the two based on individual information, and a non-invasive visual tumor evolution model is preliminarily established.

[0023] The method of mapping the evolution of chemotherapy through genomic / transcriptomic / proteomic technologies is as follows:

[0024] The paired tumor tissues collected above (responder-non-responder, before chemotherapy-optimal efficacy-disease progression) were analyzed as follows: (1) single-cell suspensions were prepared and unbiased separation of cells was achieved to obtain tumor cell populations with different drug sensitivities; (2) tumor cells were isolated and single-cell genome / transcriptome / proteome multidimensional omics detection was performed to analyze the molecular heterogeneity of tumor patients with different drug sensitivities. Bioinformatics analysis showed the evolutionary characteristics of tumor tissues of patients with different efficacy at the gene / transcription / protein levels.

[0025] Through bioinformatics analysis, we compare the evolutionary feature maps based on circulation omics, imaging omics, and genome / transcriptomics / proteomics, draw the evolutionary map of multidimensional omics, and find and fit the key regulatory pathways involved in chemotherapy drug resistance.

[0026] In a second aspect, the present invention provides a visualization diagnosis and treatment model of gastric cancer evolution driven by chemotherapy pressure, which is constructed using the method described above.

[0027] The present invention screened out differentially expressed genes through in vitro cells, which were the same as differentially expressed genes screened by other approaches, and verified through in vivo animal models.

[0028] Therefore, the third aspect of the present invention provides the application of a visualized diagnostic and therapeutic model of gastric cancer evolution driven by chemotherapy pressure, which can be used to screen gastric cancer therapeutic drugs and can also be used to evaluate the clinical value of key pathways in the evolutionary laws driven by chemotherapy drugs.

[0029] The beneficial protection and effects of the present invention are as follows:

[0030] The present invention uses a variety of bioinformatics methods to reduce the dimension and visualize the multidimensional omics data of gastric cancer, establish associations between various graphic features and phenotypes such as clinical efficacy and outcome, and comprehensively depict the evolutionary characteristics of gastric cancer chemotherapy, and explore the clonal evolution of gastric cancer cells driven by chemotherapy pressure. With "multi-point dynamic sampling" and "multi-omics combination", the goal is to establish a visualization platform that can mirror the above characteristics and integrate imaging omics, genome / transcriptomics / proteomics, molecular pathology and clinical characterization, and comprehensively display the directional clonal evolution characteristics of gastric cancer driven by chemotherapy pressure, providing decision support for more reasonable clinical layout of drug use strategies.

[0031] The model constructed according to the present invention can obtain markers or drug target protein markers (groups) that can indicate drug resistance, analyze the evolution of tumors driven by chemotherapy drug pressure, and realize real-time and accurate decision-making for clinical targeted intervention in tumor evolution and full-process anti-tumor management. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram showing the construction idea of ​​the gastric cancer evolution visualization diagnosis and treatment model of the present invention;

[0033] Figure 2 The images, tumor genome, and immune microenvironment of gastric cancer patients before and after treatment are visualized: A. Left → right: CT manifestations of primary lesions, plasma free DNA detection, and plasma exosome detection in patients with PR (partial remission) before and after treatment; B. Left → right: CT manifestations of primary lesions, plasma free DNA detection, and plasma exosome detection in patients with SD (stable disease) before and after treatment; C. Left → right: CT manifestations of primary lesions, plasma free DNA detection, and plasma exosome detection in patients with PD (progressive disease) before and after treatment;

[0034] Figure 3 The correlation diagram between five radiomic features and chemotherapy-induced disease progression of gastric cancer in 12 patients is shown, and AL represents patients with different conditions;

[0035] Figure 4The key gene mutation profiles and KEGG pathway enrichment analysis of the chemotherapy response group and chemotherapy non-response group were displayed;

[0036] Figure 5 The gene mutation spectrum analysis of 15 PR patients is shown: A. The mutant gene map of patients at baseline, disease remission and disease progression; B. KEGG pathway enrichment at baseline; C. KEGG pathway enrichment at PD;

[0037] Figure 6 The changes in gene copy number and key gene screening in different chemotherapy response groups are shown: A. SNV map of 30 patients in BL / PR / PD; B. Wilcox Test results of VEGFA gene; C. VEGFA SNV change trend of 6 representative PR patients. DETAILED DESCRIPTION

[0038] The implementation of the present invention is described in detail below in conjunction with the embodiments of the present invention. The following embodiments are implemented on the premise of the technical solution of the present invention, and provide detailed implementation methods and specific operating processes, but the protection scope of the present invention is not limited to the following embodiments.

[0039] The experimental methods in the following examples, unless otherwise specified, are all conventional methods, and are performed according to the techniques or conditions described in the literature in the field or according to the product instructions. The materials, reagents, etc. used in the following examples, unless otherwise specified, can all be obtained from commercial channels.

[0040] Figure 1 The schematic diagram of the construction idea of ​​the gastric cancer evolution visualization diagnosis and treatment model of the present invention is shown. Tumor circulation omics, imaging omics and genomics / transcriptomics / proteomics are used to depict the heterogeneity and evolutionary characteristic spectrum of gastric cancer driven by chemotherapy pressure, and a visualization evolution model is established through in vitro and in vivo functional verification and integration of clinical manifestations.

[0041] In a specific embodiment of the present invention, the construction of a gastric cancer evolution visualization diagnosis and treatment model based on a platinum-based chemotherapy regimen is taken as an example, and the specific process is as follows:

[0042] Step 1: Dynamically collect peripheral blood and tissue specimens from gastric cancer patients who are receiving platinum-based chemotherapy before chemotherapy, at the best efficacy, and at disease progression, and establish a standardized sample library

[0043] A. Dynamically collect peripheral blood from gastric cancer patients who initially use platinum-based chemotherapy drugs before chemotherapy, at the time of optimal therapeutic effect, and at the time of disease progression. Use the microfluidic cell capture instrument detection reagent Nextctc to obtain CTCs based on antibody-independent circulating tumor cell (CTC) capture and non-destructive release technology. Use two-dimensional and three-dimensional systems to culture and expand cells for subsequent functional verification and various omics tests.

[0044] B. The peripheral blood collected at the above three time points was separated into plasma, and free DNA (cfDNA), plasma exosomes and cytokine chips were detected.

[0045] C. Collect paired "responder-non-responder" and "before chemotherapy-best efficacy-disease progression" tumor tissues, perform cell separation, culture and amplify cells using two-dimensional and three-dimensional systems, in preparation for subsequent functional verification and genomic / transcriptomic / proteomic testing.

[0046] D. Establish a clinical database to collect basic information of patients (treatment status, progression time, prognosis, pathological data, imaging images and data, etc.

[0047] Step 2: Through circulation omics, imaging omics and Genomic / transcriptomic / proteomic technologies , analyze the heterogeneity of individuals with different efficacy, draw a multi-dimensional omics chemotherapy evolution map, and build a tumor clone evolution network control platform through bioinformatics analysis

[0048] A. Cyclomic evolution map:

[0049] (1) Perform single-cell sequencing on the captured cells, analyze the differences in tumor cell gene spectra before chemotherapy, at optimal efficacy, and at disease progression using bioinformatics algorithms, evaluate the direction of clonal evolution, and obtain an evolutionary gene set. (2) Separate plasma at different time points, detect cfDNA, plasma exosomes, and cytokines, find differential expression changes in different individuals and at different efficacy, draw a tumor evolution map, explore evolutionary laws, and tap potential platinum-driven tumor evolution pathways.

[0050] Figure 2 The visualization of the images, tumor genome, and immune microenvironment of gastric cancer patients before and after treatment of PR, SD, and PD after platinum-based chemotherapy is shown: A. Left to right: CT manifestations of primary lesions, plasma free DNA detection, and plasma exosome detection in patients with PR (partial remission) before and after treatment; B. Left to right: CT manifestations of primary lesions, plasma free DNA detection, and plasma exosome detection in patients with SD (stable disease) before and after treatment; C. Left to right: CT manifestations of primary lesions, plasma free DNA detection, and plasma exosome detection in patients with PD (progressive disease) before and after treatment. The CT manifestations of primary lesions, cfDNA detection, and exosome detection results of patients with different conditions were significantly different.

[0051] B. Radiomics evolutionary atlas: Using radiomics technology, we extracted the imaging data of patients before chemotherapy, at the best therapeutic effect, and at different stages of tumor progression, which were paired with blood and tissue specimens, and analyzed the imaging differences between individuals with different therapeutic effects. At the same time, combined with circulating omics data, we used bioinformatics technology to analyze the correlation between the two based on individuals, and initially established a non-invasive visual tumor evolution model.

[0052] Figure 3 The correlation diagram of 5 imaging features and chemotherapy disease progression of gastric cancer in 12 patients is shown. The 12 patients included three conditions: PR, SD, and PD. The correlation with chemotherapy disease progression of gastric cancer was observed through five imaging features: original_firstorder_Entropy, original_firstorder_InterquartileRange, original_firstorder_Energy, original_firstorder_90Percentile, and original_firstorder_10Percentile.

[0053] C. Use single-cell sequencing technology to depict the evolutionary map of genome / transcriptome / proteome: The paired tumor tissues collected above, including response-non-response, before chemotherapy-optimal efficacy-disease progression, (1) are prepared into single-cell suspensions, and microfluidics (Fluidigm C1 and 10x Genomics Chromium system) are used to achieve unbiased cell separation to obtain tumor cell populations with different drug sensitivities; (2) tumor cells are separated and multi-dimensional genomic / transcriptome / proteome detection of single cells is performed. The molecular heterogeneity of tumor patients with different drug sensitivities is analyzed, and bioinformatics analysis shows the evolutionary characteristics of tumor tissues of patients with different efficacy at the gene / transcription / protein level.

[0054] D. Use the principles of tumor evolution to identify and analyze branch mutations and trunk mutations in cancer cells, and establish a visualization platform for the interaction between tumor heterogeneity and tumor clonal evolution based on "trunk and branch variations".

[0055] E. Through bioinformatics analysis, compare the evolutionary feature maps based on circulation omics, imaging omics, and genome / transcriptome / proteomics, draw a multidimensional omics evolutionary map, find and fit the key regulatory pathways involved in platinum resistance (currently preliminarily verified as angiogenesis pathways), and construct an evolutionary pathway regulatory network.

[0056] Step 3: Verify changes in the "angiogenesis" pathway driven by chemotherapy drug pressure at the cellular and animal levels.

[0057] A. In vitro cell level validation

[0058] (1) Platinum-resistant cell lines were constructed (the cell lines used were human gastric cancer MKN45 and MGC-803, and the chemotherapy drugs were oxaliplatin and cisplatin. The cells were induced for 6 to 8 months by increasing the drug concentration. Finally, the resistance index (RI) was calculated by detecting the IC50 of the resistant cell lines. RI = IC50 of the resistant cell line / IC50 of the parental cell line. If RI>5, it is considered that the resistance of the resistant cell line meets the requirements of the resistant cell line). The expression of the angiogenesis pathway genes enriched above was verified by RNA-seq, and the three genes with the most obvious expression differences (VEGF-A, HGF, PDGF) were selected for subsequent research. (2) The above-mentioned successfully verified pathway genes were overexpressed / underexpressed by siRNA technology, and the RNA and protein levels of their upstream and downstream genes were quantitatively detected by RT-PCR and Western blot experiments.

[0059] C. In vivo animal model validation:

[0060] (1) Gastric cancer PDX and organoid models were constructed. After successful modeling, chemotherapy drugs (oxaliplatin and cisplatin) were used for treatment to obtain differential drug sensitivity phenotypes, and the expression of genes related to the above pathways was detected in tumor tissues of each phenotype. (2) After intervening in vitro validated key genes in gastric cancer cell lines, a nude mouse subcutaneous transplant tumor model was established, chemotherapy drugs were given, and the drug efficacy response of mice and the changes in key related genes in tumor tissues were compared.

[0061] Figure 4 The key gene mutation maps and KEGG pathway enrichment analysis of the chemotherapy response group and chemotherapy non-response group were displayed, and the RTK-RAS, P13K, Wnt, TGFbeta, p53, Notch, Myc, Cell-cycle, Nrf2, and Hippo metabolic pathways were obtained.

[0062] Step 4: Evaluate the clinical value of key pathways in the evolutionary laws driven by chemotherapy drugs and establish a visual tumor evolution model

[0063] A. Through sample library analysis and further collection of tissue specimens and peripheral blood from gastric cancer patients before and after treatment, the key genes of the drug resistance pathway screened out in the early stage were detected. Combined with clinical follow-up data, the correlation between the changes in the key molecules of the above signaling pathways and the efficacy of drug treatment was verified.

[0064] B. Based on further expansion of clinical sample size, a multi-omics combined approach was used to further verify the effectiveness of the above-mentioned signaling pathways in the evolutionary laws driven by chemotherapy drugs, and a dynamic visual tumor evolution model was established in combination with clinical characterization data.

[0065] Figure 5The gene mutation spectrum analysis of 15 PR patients is shown. Figure 6 The changes in gene copy number and key gene screening in different chemotherapy response groups are shown. By comparing the KEGG pathway enrichment between the two situations, differential genes can be screened out.

[0066] The embodiments of the present invention have been specifically described above, but the present invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure, characterized in that: The steps include: A. Establish a standardized sample library Dynamically collect peripheral blood from gastric cancer patients who initially use basic chemotherapy drugs before chemotherapy, at the time of best efficacy, and at the time of disease progression, obtain circulating tumor cells from them, culture and amplify them, and perform plasma testing on peripheral blood; collect paired response-non-response, before chemotherapy-best efficacy-disease progression tumor tissues, separate cells, culture and amplify them; establish a clinical database based on basic patient information, B. Constructing a tumor clone evolution network control platform Through the use of circulating omics, imaging omics, and genomic / transcriptomic / proteomic technologies, the heterogeneity of individuals with different therapeutic effects is analyzed, and a multi-dimensional omics chemotherapy evolution map is drawn; through bioinformatics analysis combined with cancer cell branch mutation and trunk mutation analysis, a tumor clone evolution network control station is constructed to find and fit the key regulatory pathways involved in chemotherapy resistance. C. Verification and critical path assessment Differentially expressed factors were screened based on the key regulatory pathways, and the clinical value of the key pathways in the evolutionary laws driven by chemotherapy drugs was evaluated after verification.

2. The method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure according to claim 1, characterized in that: in, In step A, the basic chemotherapy drug is selected from platinum chemotherapy drugs, fluorouracil chemotherapy drugs or paclitaxel chemotherapy drugs; Tumor cells obtained from peripheral blood and tumor tissues were cultured and expanded using two-dimensional and three-dimensional systems; When plasma testing is performed on peripheral blood, free DNA, plasma exosomes, and cytokines are tested; The patient's basic information includes treatment status, progression time, prognosis, pathological data, imaging images and data.

3. The method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure according to claim 1, characterized in that: in, In step B, the method of mapping the evolution of chemotherapy by cyclic omics is as follows: (1) performing single-cell sequencing on the tumor cells captured in step (A), analyzing the differences in the gene spectra of tumor cells at three time points before chemotherapy, at the best therapeutic effect, and at the time of disease progression by bioinformatics, evaluating the direction of clonal evolution, and obtaining an evolutionary gene set; (2) isolating the plasma at the three time points, detecting cfDNA, plasma exosomes, and cytokines, and looking for differential expression changes in different individuals and at different therapeutic effects, mapping the tumor evolution, exploring the evolutionary laws, and mining potential tumor evolution pathways driven by chemotherapy drugs.

4. The method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure according to claim 1, characterized in that: in, In step B, the method of drawing the chemotherapy evolution map through radiomics is as follows: using radiomics technology, extract the imaging data of patients paired with blood and tissue specimens before chemotherapy, at the best therapeutic effect and at different stages of tumor progression, and analyze the imaging differences between individuals with different therapeutic effects; at the same time, combined with circulating omics data, use bioinformatics technology to analyze the correlation between the two, and preliminarily establish a non-invasive visual tumor evolution model.

5. The method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure according to claim 1, characterized in that: in, In step B, the method of mapping the evolution of chemotherapy using genomic / transcriptomic / proteomic technology is as follows: the paired tumor tissues collected above at the time of response-non-response, before chemotherapy-optimal efficacy-disease progression are analyzed as follows: (1) single-cell suspensions are prepared and unbiased separation of cells is achieved to obtain tumor cell populations with different drug sensitivities; (2) tumor cells are isolated and single-cell genomic / transcriptomic / proteomic multidimensional omics detection is performed to analyze the molecular heterogeneity of tumor patients with different drug sensitivities, and bioinformatics analysis is used to show the evolutionary characteristics of tumor tissues of patients with different efficacy at the gene / transcription / protein levels.

6. The method for constructing a gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure according to claim 1, characterized in that: in, In step B, through bioinformatics analysis, the evolutionary feature maps based on circulation omics, imaging omics, and genome / transcriptomics / proteomics are compared to draw a multidimensional omics evolutionary map, and find and fit the key regulatory pathways involved in chemotherapy drug resistance.

7. A gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure, characterized in that: The method is constructed by any one of claims 1 to 6.

8. Application of the gastric cancer evolution visualization diagnosis and treatment model driven by chemotherapy pressure as described in claim 7 in the screening of gastric cancer therapeutic drugs.

Images