A method for constructing a platinum drug sensitivity prediction model and application thereof

Abstract

The application relates to the technical field of drug sensitivity prediction, and discloses a construction method of a platinum drug sensitivity prediction model and application thereof, which comprises the following steps: constructing a training set; according to the training set, a plurality of candidate models are trained through a plurality of machine learning algorithms respectively; and according to preset model indexes, the plurality of candidate models are screened to obtain at least two platinum drug sensitivity prediction models. Based on cancer type data of a target group, administered platinum drug type data, immune cell infiltration level data of tumor sample tissue and PTC data, a plurality of high-precision platinum drug sensitivity prediction models are trained by using a machine learning algorithm, and the platinum drug sensitivity prediction models can be used in combination with a hard voting method to realize accurate prediction of platinum drug sensitivity. The model has high prediction accuracy, is more targeted for the applicable population, has simple and clear prediction results, has a short use cycle and is low in cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of drug sensitivity prediction, in particular to a construction method of a platinum drug sensitivity prediction model and application thereof. BACKGROUND

[0002] At present, the chemotherapy regimen for cancer is mainly selected according to the clinical stage and pathological type of cancer. However, due to the heterogeneity of tumors and the insensitivity of tumors to chemotherapy drugs, the treatment is blind. The chemotherapy drugs themselves have cytotoxicity, and during the chemotherapy process, the cancer may not be controlled, but adverse reactions may occur, causing unnecessary pain to patients, affecting the quality of life, and even affecting the prognosis of tumors. For example, platinum chemotherapy drugs, including cisplatin, carboplatin, lobaplatin and oxaliplatin, can treat gastrointestinal tumors, gynecological disease tumors, lung cancer and other diseases. Although this type of drug can treat tumors, due to the blind drug regimen, it will also produce side effects such as liver and kidney function damage and bone marrow suppression.

[0003] With the rapid development of molecular biology, it is found that the gene expression in cancer is closely related to the occurrence, development, metastasis and prognosis of tumors, and has a certain connection with the efficacy of tumor chemotherapy. Moreover, more and more machine learning models are used to predict drug sensitivity. How to use machine learning methods to accurately predict patient drug sensitivity and assist in guiding individualized treatment of tumor patients has important clinical significance. At present, most drug sensitivity prediction models use data mainly from public databases, which lack real clinical data for Chinese people, making the prediction results uneven compared with the real clinical effect, and the prediction results of the model are IC50 values, which cannot intuitively show the drug effect.

[0004] Therefore, there is an urgent need for a construction method of a platinum drug sensitivity prediction model to assist in intuitively and accurately predicting patient drug sensitivity, guiding individualized treatment of tumor patients, and improving patient recovery rate. SUMMARY

[0005] The present application provides a construction method of a platinum drug sensitivity prediction model and application thereof, which solves the defects of the existing drug sensitivity prediction model that the prediction result accuracy is low, and the prediction result is IC50 value, which cannot intuitively show the drug effect.

[0006] The present application provides a construction method of a training set, comprising:

[0007] Obtaining whole transcriptome sequencing data, cancer type data, administered platinum drug type data and platinum drug sensitivity data of tumor sample tissues of a target population, wherein the target population comprises a cancer patient population, and the platinum drug sensitivity data comprises platinum drug response data and platinum drug resistance data;

[0008] obtaining gene expression data of the tumor sample tissue of the target population according to the whole transcriptome sequencing data of the tumor sample tissue of the target population;

[0009] obtaining gene expression data of the tumor sample tissue of the target population according to the whole transcriptome sequencing data of the tumor sample tissue of the target population;

[0010] obtaining gene expression data of the tumor sample tissue of the target population according to the whole transcriptome sequencing data of the tumor sample tissue of the target population;

[0011] In an embodiment, the type of platinum drug administered data includes any one or any combination of the following: cisplatin data, carboplatin data, lobaplatin data, oxaliplatin data.

[0012] In an embodiment, the obtaining gene expression data of the tumor sample tissue of the target population according to the whole transcriptome sequencing data of the tumor sample tissue of the target population comprises:

[0013] obtaining reference genome data;

[0014] performing data quality control on the whole transcriptome sequencing data of the tumor sample tissue of the target population;

[0015] aligning the whole transcriptome sequencing data after data quality control with the reference genome data to obtain a bam file;

[0016] calculating gene expression according to the bam file to obtain gene expression data of the tumor sample tissue of the target population.

[0017] In an embodiment, the performing data quality control on the whole transcriptome sequencing data of the tumor sample tissue of the target population comprises:

[0018] performing data quality control on the whole transcriptome sequencing data of the tumor sample tissue of the target population according to preset data quality control conditions, wherein the preset data quality control conditions comprise:

[0019] filtering out reads with a content of base N exceeding a preset content threshold;

[0020] filtering out reads with a sequencing quality value lower than a preset quality value threshold.

[0021] In an embodiment, the preset quality value threshold is 15.

[0022] In an embodiment, the obtaining gene expression data of the tumor sample tissue of the target population according to the whole transcriptome sequencing data of the tumor sample tissue of the target population comprises:

[0023] According to the gene expression data of the tumor sample tissue of the target population, a plurality of immune cell infiltration level evaluation results are obtained by a plurality of cell infiltration level evaluation methods, and then the immune cell infiltration level data of the tumor sample tissue of the target population is obtained, wherein the plurality of cell infiltration level evaluation methods include any one or any combination thereof: CIBERSORT, CIBERSORT-ABS, EPIC, MCPCOUNTER, QUANTISEQ, TIMER, XCELL.

[0024] In an embodiment, the present application provides a training set construction method, which further comprises:

[0025] According to the preset cleaning rule, the training set is data cleaned, wherein the preset cleaning rule includes any one or any combination thereof:

[0026] For each piece of immune cell infiltration level data in the immune cell infiltration level data of the tumor sample tissue of the target population, when > 60% of the immune cell infiltration level evaluation results are 0, the piece of immune cell infiltration level data is deleted;

[0027] For the immune cell infiltration level data of the tumor sample tissue of the target population obtained by a plurality of cell infiltration level evaluation methods, when > 80% of the values in the immune cell infiltration level evaluation results obtained by a certain cell infiltration level evaluation method are 0, the immune cell infiltration level evaluation results obtained by the cell infiltration level evaluation method are deleted;

[0028] When the cancer data is represented as the same cancer, and the type of platinum drug administered is represented as the same platinum drug, for the platinum drug sensitivity data of the target population, if the ratio of platinum drug response and platinum drug resistance represented by the platinum drug sensitivity data is outside the preset ratio range, then part of the platinum drug sensitivity data is randomly deleted until the ratio of platinum drug response and platinum drug resistance is within the preset range.

[0029] The present application provides a platinum drug sensitivity prediction model construction method, comprising:

[0030] Obtaining the training set obtained by the training set construction method according to any one of the above;

[0031] According to the training set, a plurality of candidate models are trained by a plurality of machine learning algorithms, respectively;

[0032] According to the preset model index, the plurality of candidate models are screened to obtain at least two final candidate models as at least two platinum drug sensitivity prediction models.

[0033] In an embodiment, the at least two final candidate models include four final candidate models, wherein the four final candidate models are candidate models trained by Decision Tree, Random Forest, Adaboost, and xgboost machine learning algorithms, respectively, according to the training set.

[0034] In an embodiment, the preset model indicators include any one or any combination thereof of: Area Under Curve (AUC), F1-score, Accuracy, and Matthews Correlation Coefficient (MCC).

[0035] The present application provides a platinum drug sensitivity prediction system, comprising:

[0036] The data receiving module is configured to receive cancer type data of a to-be-tested person, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue sent by at least one client, wherein the to-be-tested person is a cancer patient.

[0037] The prediction module is configured to obtain platinum drug sensitivity prediction results of the to-be-tested person by using at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model according to the cancer type data of the to-be-tested person, the platinum drug type data to be administered, and the immune cell infiltration level data of the tumor sample tissue, and combining a hard voting method.

[0038] The data sending module is configured to send the platinum drug sensitivity prediction results of the to-be-tested person back to the at least one client.

[0039] In an embodiment, the prediction module includes:

[0040] The prediction submodule is configured to obtain at least two platinum drug sensitivity prediction results of the to-be-tested person by using at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model according to the cancer type data of the to-be-tested person, the platinum drug type data to be administered, and the immune cell infiltration level data of the tumor sample tissue, respectively.

[0041] The determination submodule is used to: obtain the final platinum drug sensitivity prediction result of the test subject based on at least two platinum drug sensitivity prediction results and a hard voting method. The hard voting method is as follows: when there is a unique platinum drug sensitivity prediction result with the most votes among the at least two platinum drug sensitivity prediction results of the test subject, the unique platinum drug sensitivity prediction result with the most votes is taken as the final platinum drug sensitivity prediction result of the test subject; when there is no unique platinum drug sensitivity prediction result with the most votes among the at least two platinum drug sensitivity prediction results of the test subject, the platinum drug sensitivity prediction result of the platinum drug sensitivity prediction model with the highest platinum drug sensitivity prediction accuracy corresponding to the platinum drug type is taken as the final platinum drug sensitivity prediction result of the test subject.

[0042] The present invention also provides an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the computer program to implement the training set construction method and the platinum drug sensitivity prediction model construction method described above.

[0043] The present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the training set construction method and the platinum-based drug sensitivity prediction model construction method described above.

[0044] The present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute any of the above-described methods for constructing a training set and for constructing a platinum-based drug sensitivity prediction model.

[0045] This invention provides a method for constructing a platinum-based drug sensitivity prediction model and its application. Based on cancer type data of the target population, data on the type of platinum-based drug administered, data on the level of immune cell infiltration in tumor sample tissue, and platinum-based drug sensitivity data (PTC data), multiple high-precision platinum-based drug sensitivity prediction models are trained using machine learning algorithms. These models can be used in conjunction with hard voting methods to achieve accurate prediction of platinum-based drug sensitivity. The platinum-based drug sensitivity prediction model constructed in this invention has high prediction accuracy, with an average accuracy greater than 80%; it is more targeted to the Chinese cancer patient population; the prediction results are simple and clear, divided into platinum-based drug response and platinum-based drug resistance; the usage cycle is short, with prediction results available within 7 days (including sequencing, analysis, and prediction) from obtaining tumor sample tissue from cancer patients; and the cost is low, with the main cost being the decreasing cost of next-generation sequencing year by year. Attached Figure Description

[0046] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0047] Figure 1 A flowchart of the construction method of the platinum drug sensitivity prediction model provided by the present application.

[0048] Figure 2 A structural diagram of the platinum drug sensitivity prediction system provided by the present application.

[0049] Figure 3 A structural diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0050] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. They should not be understood as a limitation to the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application. In the description of the present application, it should be understood that the terms used are only for the purpose of description, and should not be understood as indicating or implying relative importance.

[0051] The construction method of the platinum drug sensitivity prediction model provided by the present application and its application will be described below. Figures 1-3 The construction method of the platinum drug sensitivity prediction model provided by the present application and its application will be described below.

[0052] Figure 1 The flowchart of the construction method of the platinum drug sensitivity prediction model provided by the present application is shown in FIG. 1. Referring to FIG. 1, Figure 1 The construction method of the platinum drug sensitivity prediction model provided by the present application can include:

[0053] In step S110, a training set is constructed, wherein the cancer data of the target group, the type data of the platinum drug administered, and the immune cell infiltration level data of the tumor sample tissue are used as sample data, and the platinum drug sensitivity data of the target group is used as label data.

[0054] In this embodiment, step S110 can include:

[0055] S1101, obtaining whole transcriptome sequencing data of tumor sample tissues of a target population, cancer type data (the types of cancer can include gastric cancer, colorectal cancer, ovarian cancer, breast cancer, non-small cell lung cancer, oral cancer), administered platinum drug type data, and platinum drug sensitivity data, wherein the target population includes a cancer patient population, the platinum drug sensitivity data includes platinum drug response data and platinum drug resistance data, the administered platinum drug type data includes any one or any combination thereof: cisplatin data, carboplatin data, lobaplatin data, oxaliplatin data, and the platinum drug sensitivity data is detected based on a microtumor drug sensitivity detection method (PTC) on the tumor sample tissues of the target population.

[0056] S1102, obtaining gene expression data of the tumor sample tissues of the target population according to the whole transcriptome sequencing data of the tumor sample tissues of the target population. Specifically, this step can first perform data quality control on the whole transcriptome sequencing data of the tumor sample tissues of the target population to filter out reads with a base N content exceeding a preset content threshold and filter out reads with a sequencing quality value below a preset quality value threshold (e.g., 15), then align the data quality controlled whole transcriptome sequencing data with reference genome data to obtain a bam file, and then calculate gene expression (TPM) according to the bam file to obtain the gene expression data of the tumor sample tissues of the target population.

[0057] S1103, obtaining a plurality of immune cell infiltration level evaluation results by a plurality of cell infiltration level evaluation methods according to the gene expression data of the tumor sample tissues of the target population, and then obtaining immune cell infiltration level data of the tumor sample tissues of the target population, wherein the plurality of cell infiltration level evaluation methods include CIBERSORT, CIBERSORT-ABS, EPIC, MCPCOUNTER, QUANTISEQ, TIMER, and XCELL.

[0058] In this embodiment, the gene expression data (TPM) of the tumor sample tissues of the target population is evaluated by seven cell infiltration level evaluation methods to obtain seven immune cell infiltration level evaluation results of the tumor sample tissues of the target population, and then the seven immune cell infiltration level evaluation results are combined to obtain the final immune cell infiltration level evaluation result of the tumor sample tissues of the target population as the immune cell infiltration level data of the tumor sample tissues of the target population.

[0059] S1104, performing data cleaning on the training set according to a preset cleaning rule, wherein the preset cleaning rule includes:

[0060] For each of the immune cell infiltration level data of the tumor sample tissue of the target population, when the value of >60% of the immune cell infiltration level evaluation results is 0, the immune cell infiltration level data is deleted (because one piece of immune cell infiltration level data includes seven immune cell infiltration level evaluation results, i.e., when the value of >60% of the immune cell infiltration level evaluation results is 0, the piece of immune cell infiltration level data is deleted).

[0061] For the immune cell infiltration level data of the tumor sample tissue of the target population obtained by the plurality of cell infiltration level evaluation methods, when the value of >80% of the immune cell infiltration level evaluation results obtained by a certain cell infiltration level evaluation method is 0, the immune cell infiltration level evaluation results obtained by the cell infiltration level evaluation method are deleted (i.e., for all immune cell infiltration level evaluation results, if the value of >80% of the immune cell infiltration level evaluation results obtained by a certain cell infiltration level evaluation method is 0, the immune cell infiltration level evaluation results of the tumor sample tissue obtained by the cell infiltration level evaluation method are deleted).

[0062] When the cancer type data of the target population and the type of platinum drug administered are the same, for the platinum drug sensitivity data of the target population, if the ratio of platinum drug response and platinum drug resistance indicated by the platinum drug sensitivity data is outside the preset ratio range, part of the platinum drug sensitivity data is randomly deleted until the ratio of platinum drug response and platinum drug resistance is within the preset range.

[0063] S1105, using the cancer type data of the target population, the type of platinum drug administered, and the immune cell infiltration level data of the tumor sample tissue as sample data, and using the platinum drug sensitivity data of the target population as label data, a training set is constructed.

[0064] Step S120, according to the training set, a plurality of candidate models are trained by a plurality of machine learning algorithms respectively. The plurality of machine learning algorithms can include machine learning algorithms with classification tasks, such as decision tree, naive Bayes, logistic regression, K-nearest neighbor, support vector machine, random forest, adaptive boosting, gradient boosting tree algorithm, etc.

[0065] Step S130, screening the plurality of candidate models according to a preset model index to obtain at least two final candidate models as at least two platinum drug sensitivity prediction models, wherein the preset model index includes any one or any combination thereof: area under the ROC curve (AUC), F1 score, accuracy, Matthews correlation coefficient (MCC).

[0066] In this embodiment, four final candidate models are obtained, which are candidate models trained by decision tree (Decision Tree), random forest (RandomForest), adaptive boosting (Adaboost) and gradient boosting tree algorithm (xgboost) machine learning algorithms. Four platinum drug sensitivity prediction models are obtained in this embodiment.

[0067] The decision tree (Decision Tree) machine learning algorithm constructs a tree structure through a series of feature-based branch nodes. Starting from the root node, the data (including cancer data of the target population, type data of the administered platinum drugs, and immune cell infiltration level data of the tumor sample tissue) is gradually divided into different child nodes according to different values of the features, until the leaf node is reached. Each leaf node corresponds to a class (platinum drug response or platinum drug resistance). It is easy to understand and interpret, can process multiple types of data (cancer data of the target population, type data of the administered platinum drugs, and immune cell infiltration level data of the tumor sample tissue), including numerical and categorical types, and has relatively high computational efficiency.

[0068] The random forest (RandomForest) machine learning algorithm is an ensemble learning algorithm that makes prediction classification based on the combination of multiple decision trees. It has high accuracy and stability, can effectively avoid overfitting, can process high-dimensional data including cancer data of the target population, type data of the administered platinum drugs, and immune cell infiltration level data of the tumor sample tissue, is not sensitive to feature selection, and can evaluate the importance of features.

[0069] The adaptive boosting (Adaboost) machine learning algorithm initializes the weight distribution of the data in the training set, so that each tumor sample tissue data has the same weight. In each iteration, a weak learner is trained, and the error of the weak learner on the training set is calculated. The weights are updated according to the error, and multiple weak learners are combined to form a strong learner. It can improve the performance of weak learners, is not prone to overfitting, and has certain robustness to outliers and noise data.

[0070] The core principle of the XGBoost machine learning algorithm is the gradient boosting tree algorithm, which continuously adds new trees to fit the residual error of the previous round of model, and gradually optimizes the objective function (composed of loss function and regularization term); wherein, in the process of constructing the tree, the exact greedy algorithm or approximate algorithm is used to select the optimal split node to maximize the gain after splitting. It is efficient, flexible and accurate.

[0071] After training the platinum drug sensitivity prediction model, it can be applied to the platinum drug sensitivity prediction system to obtain the platinum drug sensitivity prediction results of the to-be-tested person to different platinum drugs according to the cancer data of the to-be-tested person, the type data of the to-be-administered platinum drug (the type of the to-be-administered platinum drug can be multiple), the immune cell infiltration level data of the tumor sample tissue, and combined with the hard voting method, to assist in intuitively and accurately predicting drug sensitivity of patients, guiding personalized treatment of tumor patients, and improving the recovery rate of patients.

[0072] The construction and application of the platinum drug sensitivity prediction model will be described in two embodiments as follows.

[0073] Example 1, construction of platinum drug sensitivity prediction model

[0074] Example 1 is directed to the tumor tissue samples of 1021 cancer patients (target population) participating in modeling, and the operation steps include:

[0075] 1. Using the NovaSeq X Plus platform of Illumina Company to perform LncRNA (whole transcriptome) sequencing, the sequencing data amount is 30G per sample, and the whole transcriptome sequencing data of the tumor sample tissue of the target population is obtained.

[0076] 2. Bioinformatics analysis of sequencing samples, using Fastp (v 0.23.2) to perform data quality control on the whole transcriptome sequencing data of the tumor sample tissue of the target population to generate cleandata (data quality controlled whole transcriptome sequencing data), and using STAR software (v 2.7.10a) to align the cleandata data with the reference genome data (GRCh38) to obtain bam files.

[0077] 3. Using RSEM (v 1.3.3) to calculate gene expression by the generated bam file to obtain TPM value, i.e. to obtain the gene expression data of the tumor sample tissue of the target population.

[0078] 4. Using the gene expression data of the tumor sample tissue of the target population and seven cell infiltration level evaluation methods (CIBERSORT, CIBERSORT-ABS, EPIC, MCPCOUNTER, QUANTISEQ, TIMER, XCELL) to evaluate the immune cell infiltration level of the tumor sample tissue of the target population, and obtain the immune cell infiltration level data of the tumor sample tissue of the target population.

[0079] 5. The PTC drug sensitivity response of the cancer patient is classified into two categories, namely platinum drug response (0 represents) and platinum drug resistance (1 represents).

[0080] 6. The integrated data set includes: cancer patient number, platinum drug type number, cancer type number, immune cell infiltration level of tumor sample tissue evaluated by different methods, and platinum drug sensitivity classification.

[0081] 7. By dividing the data set into 70% training set and 30% test set, 10-fold cross-validation, using AUC and F1_score indicators to select four final candidate models as four platinum drug sensitivity prediction models.

[0082] The optimal key parameters of the model set in Example 1 when training the four final candidate models are shown in Table 1.

[0083] Table 1 Optimal key parameters of the model

[0084]

[0085]

[0086] The training results of the final candidate model of Example 1 are shown in Table 2.

[0087] Table 2 Model training results

[0088]

[0089] After determining the optimal key parameters of the model, Example 1 uses the test set to verify the generalization ability of each final candidate model, and the model test accuracy is shown in Table 3.

[0090] Table 3 Model test accuracy

[0091]

[0092] Example 2, using the platinum drug sensitivity prediction model constructed in Example 1 to predict the platinum drug sensitivity of 209 cancer patients to platinum drugs (cisplatin, carboplatin, lobaplatin, oxaliplatin).

[0093] The specific steps include:

[0094] 1. Obtain LncRNA sequencing data (i.e. whole transcriptome sequencing data) of tumor sample tissues of 209 cancer patients. The sample sequencing amount is guaranteed to be at least 30G, and the 209 cancer patients are all single-drug, which are respectively: 84 patients have clinical drug response data, and the drug response types are divided according to the tag classification; 125 patients are drug sensitivity values measured by PTC, and the drug response types are divided according to the drug sensitivity data.

[0095] 2. Bioinformatics analysis of the whole transcriptome sequencing data in 1, including data quality control, reference genome data alignment, and quantitative gene expression.

[0096] 3. Using the gene expression data of the tumor sample tissues of the 209 cancer patients, seven methods (CIBERSORT, CIBERSORT-ABS, EPIC, MCPCOUNTER, QUANTISEQ, TIMER, XCELL) are used to evaluate the immune cell infiltration level of the tumor sample tissues.

[0097] 4. Generate model input data d1, which contains cancer patient number, platinum drug type number to be administered, cancer type number, and immune cell infiltration level of tumor sample tissues evaluated by different methods.

[0098] 5. Input the model input data d1 into the four platinum drug sensitivity prediction models constructed in Example 1, and combine the hard voting method (integrate the platinum drug sensitivity prediction results of the four platinum drug sensitivity prediction models, the majority class voting result is the final result, if it is a tie, select the result of the model with the highest prediction accuracy of the corresponding drug in the test data as the final result) to get the platinum drug sensitivity prediction results of the 209 cancer patients.

[0099] 6. Consistency comparison of the platinum drug sensitivity prediction results of the 209 cancer patients and their true drug response types, and the comparison results of the clinical samples and PTC samples are shown in Tables 4 and 5, respectively:

[0100] Table 4 Consistency comparison of clinical samples

[0101]

[0102] Table 5 Consistency comparison of PTC samples

[0103]

[0104] The accuracy of the platinum drug sensitivity prediction model is predicted by 209 clinical data and PTC drug sensitivity data. After comparison, the platinum drug prediction effect (cisplatin, carboplatin, lobaplatin, oxaliplatin) is better, wherein, in the clinical sample, the cisplatin accuracy is 82.8%, the carboplatin accuracy is 90%, the lobaplatin accuracy is 87.5%, and the oxaliplatin accuracy is 80.95%; In patients with PTC detection, the cisplatin accuracy is 86.35%, the carboplatin accuracy is 89.5%, the lobaplatin accuracy is 85.75%, and the oxaliplatin accuracy is 82.25%.

[0105] The application provides a platinum drug sensitivity prediction model construction method and application thereof, and at least has the following

[0106] Beneficial effects:

[0107] (1) high precision;

[0108] From the results of Table 4 and Table 5, whether consistent with clinical drug data or PTC obtained drug sensitivity data, the platinum drug sensitivity prediction model constructed by the application has very high prediction accuracy (>80%).

[0109] (2) short cycle, more economical and efficient;

[0110] From the acquisition of tumor sample tissue of cancer patients, the prediction result can be obtained within 7 days (including sequencing, analysis and prediction), and it takes 2 weeks to obtain drug sensitivity data through PTC; as the sequencing cost decreases year by year, the cost of drug sensitivity detection of patients based on gene expression data will also decrease year by year.

[0111] (3) The modeling data uses clinical data specially for Chinese people, and has race specificity and high clinical effect consistency.

[0112] (4) The prediction result is simple and clear, and is divided into response and drug resistance, which is convenient for use as auxiliary drug guidance.

[0113] The platinum drug sensitivity prediction system provided by the application is described below, and the platinum drug sensitivity prediction system described below can be correspondingly referred to the platinum drug sensitivity prediction method described above.

[0114] Referring to Figure 2 The platinum drug sensitivity prediction system provided by the application can comprise:

[0115] The data receiving module is used for receiving the cancer type data of the to-be-tested person, the platinum drug type data to be applied, and the immune cell infiltration level data of the tumor sample tissue sent by at least one client, wherein the to-be-tested person is a cancer patient.

[0116] The prediction module is used to: obtain the platinum drug sensitivity prediction result of the test subject by combining at least two platinum drug sensitivity prediction models obtained by the platinum drug sensitivity prediction model construction method described above with the hard voting method, based on the cancer type data of the test subject, the type data of the platinum drug to be administered, and the immune cell infiltration level data of the tumor sample tissue.

[0117] The data transmission module is used to send the predicted results of the platinum drug sensitivity of the test subject back to at least one client.

[0118] In one implementation, the prediction module includes:

[0119] The prediction submodule is used to: obtain at least two platinum drug sensitivity prediction results for the test subject based on the cancer type data of the test subject, the type data of the platinum drug to be administered, and the immune cell infiltration level data of the tumor sample tissue, respectively, through at least two platinum drug sensitivity prediction models obtained by the platinum drug sensitivity prediction model construction method described above.

[0120] The determination submodule is used to: obtain the final platinum drug sensitivity prediction result of the test subject based on at least two platinum drug sensitivity prediction results and a hard voting method. The hard voting method is as follows: when there is a unique platinum drug sensitivity prediction result with the most votes among the at least two platinum drug sensitivity prediction results of the test subject, the unique platinum drug sensitivity prediction result with the most votes is taken as the final platinum drug sensitivity prediction result of the test subject; when there is no unique platinum drug sensitivity prediction result with the most votes among the at least two platinum drug sensitivity prediction results of the test subject, the platinum drug sensitivity prediction result of the platinum drug sensitivity prediction model with the highest platinum drug sensitivity prediction accuracy for the corresponding platinum drug type is taken as the final platinum drug sensitivity prediction result of the test subject.

[0121] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute the following steps:

[0122] receive cancer type data of a to-be-tested person, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue, wherein the to-be-tested person is a cancer patient;

[0123] obtain a platinum drug sensitivity prediction result of the to-be-tested person according to the cancer type data of the to-be-tested person, the platinum drug type data to be administered, the immune cell infiltration level data of the tumor sample tissue, and at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model according to any one of the preceding embodiments, in combination with a hard voting method;

[0124] send the platinum drug sensitivity prediction result of the to-be-tested person back to the at least one client.

[0125] In addition, the logic instructions in the memory 830 described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0126] On the other hand, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor, so that the computer can execute the following steps:

[0127] receive cancer type data of a to-be-tested person, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue, wherein the to-be-tested person is a cancer patient;

[0128] obtain a platinum drug sensitivity prediction result of the to-be-tested person according to the cancer type data of the to-be-tested person, the platinum drug type data to be administered, the immune cell infiltration level data of the tumor sample tissue, and at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model according to any one of the preceding embodiments, in combination with a hard voting method;

[0129] send the platinum drug sensitivity prediction result of the to-be-tested person back to the at least one client.

[0130] In yet another aspect, the present application also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the following steps:

[0131] receiving cancer type data of a to-be-tested person, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue, wherein the to-be-tested person is a cancer patient;

[0132] According to the cancer type data of the to-be-tested person, the platinum drug type data to be administered, and the immune cell infiltration level data of the tumor sample tissue, at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model are combined with the hard voting method to obtain the platinum drug sensitivity prediction result of the to-be-tested person.

[0133] The platinum drug sensitivity prediction result of the to-be-tested person is sent back to the at least one client.

[0134] The device embodiments described above are only schematic, wherein the units shown as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0135] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software plus necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part that makes a contribution to the prior art, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0136] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a training set, comprising: obtaining whole transcriptome sequencing data of tumor sample tissues of a target population, cancer type data, administered platinum drug type data, and platinum drug sensitivity data, wherein the target population comprises a cancer patient population, and the platinum drug sensitivity data comprises platinum drug response data and platinum drug resistance data; obtaining gene expression data of the tumor sample tissues of the target population according to the whole transcriptome sequencing data of the tumor sample tissues of the target population; obtaining immune cell infiltration level data of the tumor sample tissues of the target population according to the gene expression data of the tumor sample tissues of the target population; constructing a training set using the cancer type data, the administered platinum drug type data, and the immune cell infiltration level data of the tumor sample tissues of the target population as sample data, and using the platinum drug sensitivity data of the target population as label data.

2. The method of claim 1, wherein, The method further comprises: obtaining reference genome data; performing data quality control on the whole transcriptome sequencing data of the tumor sample tissues of the target population; aligning the data quality controlled whole transcriptome sequencing data with the reference genome data to obtain a bam file; calculating gene expression quantity according to the bam file to obtain the gene expression data of the tumor sample tissues of the target population.

3. The method of claim 2, wherein the training set is constructed by: The method further comprises: performing data quality control on the whole transcriptome sequencing data of the tumor sample tissues of the target population according to preset data quality control conditions, wherein the preset data quality control conditions comprise: filtering out reads with a content of base N exceeding a preset content threshold; filtering out reads with a sequencing quality value lower than a preset quality value threshold.

4. The method of claim 3, wherein the training set is constructed by, The method further comprises: obtaining a plurality of immune cell infiltration level evaluation results by a plurality of immune cell infiltration level evaluation methods according to the gene expression data of the tumor sample tissues of the target population, and then obtaining the immune cell infiltration level data of the tumor sample tissues of the target population, wherein the plurality of immune cell infiltration level evaluation methods comprise any one or any combination of the following: CIBERSORT, CIBERSORT-ABS, EPIC, MCPCOUNTER, QUANTISEQ, TIMER, and XCELL.

5. The method of claim 4, wherein the training set is constructed by, The method further comprises: performing data cleaning on the training set according to preset cleaning rules, wherein the preset cleaning rules comprise any one or any combination of the following: for each piece of immune cell infiltration level data in the immune cell infiltration level data of the tumor sample tissues of the target population, if the value of >60% of immune cell infiltration level evaluation results is 0, then the piece of immune cell infiltration level data is deleted. For the immune cell infiltration level data of the tumor sample tissue of the target population obtained by multiple cell infiltration level evaluation methods, when ≥80% of the values in the immune cell infiltration level evaluation results obtained by a certain cell infiltration level evaluation method are 0, the immune cell infiltration level evaluation results obtained by the cell infiltration level evaluation method are deleted; In the case that the cancer type data and the platinum drug type data are the same, if the ratio of platinum drug response and platinum drug resistance of the platinum drug sensitivity data of the target population is outside the preset ratio range, part of the platinum drug sensitivity data is randomly deleted until the ratio of platinum drug response and platinum drug resistance is within the preset range.

6. A method for constructing a platinum drug sensitivity prediction model, characterized by, The method comprises the following steps: obtaining the training set obtained by the construction method of the training set according to any one of claims 1-5; training a plurality of candidate models by using a plurality of machine learning algorithms respectively according to the training set; screening the plurality of candidate models according to a preset model index to obtain at least two final candidate models as at least two platinum drug sensitivity prediction models.

7. The method for constructing a platinum-based drug sensitivity prediction model according to claim 6, characterized in that, The at least two final candidate models include four final candidate models, wherein the four final candidate models are candidate models obtained by training a decision tree, a random forest, an adaptive boosting, and a boosting tree model machine learning algorithm respectively according to the training set.

8. A platinum drug sensitivity prediction system, characterized by, The method comprises the following steps: a data receiving module, configured to receive cancer type data, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue sent by at least one client, wherein the subject to be tested is a cancer patient; a prediction module, configured to obtain a platinum drug sensitivity prediction result of the subject to be tested by using at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model according to the cancer type data, the platinum drug type data to be administered, and the immune cell infiltration level data of the tumor sample tissue of the subject to be tested, and combining a hard voting method; a data sending module, configured to send the platinum drug sensitivity prediction result of the subject to be tested back to the at least one client.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor performs the following steps: receiving cancer type data, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue sent by at least one client, wherein the subject to be tested is a cancer patient; obtaining a platinum drug sensitivity prediction result of the subject to be tested by using at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model according to the cancer type data, the platinum drug type data to be administered, and the immune cell infiltration level data of the tumor sample tissue of the subject to be tested, and combining a hard voting method; sending the platinum drug sensitivity prediction result of the subject to be tested back to the at least one client. 10.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to perform the following steps: receiving cancer type data, platinum drug type data to be administered, and immune cell infiltration level data of a tumor sample tissue sent by at least one client, wherein the subject to be tested is a cancer patient; According to the cancer type data of the to-be-tested person, the type data of the to-be-administered platinum drug, the immune cell infiltration level data of the tumor sample tissue, at least two platinum drug sensitivity prediction models obtained by the construction method of the platinum drug sensitivity prediction model in claim 6 or 7, and the hard voting method, the platinum drug sensitivity prediction result of the to-be-tested person is obtained; The platinum drug sensitivity prediction result of the to-be-tested person is sent back to at least one client.

Images