Cancer staging method and device and electronic equipment

Abstract

According to the cancer staging method and device and the electronic equipment, the cancer staging corresponding to the input methylation data set is predicted through the staging prediction model obtained by integrating the multiple classification models, and the universality of the cancer staging is improved while the prediction accuracy is guaranteed. The method comprises the following steps: acquiring a methylation data set of a target object; inputting the methylation data set into a staging prediction model, and outputting a cancer staging value of the target object; wherein the staging prediction model is obtained by integrating n classification models, model parameters of the n classification models are obtained by optimizing through a grid retrieval method, and the n classification models are obtained by screening m classification models through a cross validation method, the m classification models are obtained by performing independent training by using the same data set, m is greater than n, m is an integer greater than 2, and n is an integer greater than 1.

Description

Technical Field

[0001] The present disclosure relates to the field of biological information technology, and in particular to a cancer staging method, device and electronic equipment. Background Art

[0002] Cancer staging is a method to determine the extent of cancer development and spread. Accurate cancer staging is conducive to formulating the most reasonable treatment plan for cancer patients and effectively judging the prognosis of cancer.

[0003] In the prior art, cancer staging is usually performed using methods such as clinical and pathological examinations. However, clinical examinations (blood tests, radiological examinations, endoscopic examinations, etc.) alone provide limited information on cancer staging and have limitations. Although pathological examinations can more accurately determine the staging, surgery is required to obtain pathological sections, and not all tumors require surgical treatment. In addition, some patients may have undergone chemotherapy and radiotherapy before pathological examinations, which may underestimate the true staging of the tumor to a certain extent. Therefore, the method of using pathological examinations still has limitations. In addition, different cancers have different staging systems, and some cancers do not have suitable staging methods, so conventional staging methods are not universal. Summary of the invention

[0004] The present disclosure provides a cancer staging method, device and electronic device, which are used to predict the cancer stage corresponding to an input methylation data set through a staging prediction model obtained by integrating multiple classification models, thereby ensuring the prediction accuracy while improving the universality of cancer staging.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for staging cancer, the method comprising:

[0006] Obtain a methylation dataset of a target object;

[0007] Inputting the methylation dataset into a staging prediction model, and outputting a cancer staging value of the target object;

[0008] Among them, the stage prediction model is obtained by integrating n classification models, the model parameters of the n classification models are optimized by using the grid retrieval method, the n classification models are screened from m classification models by using the cross-validation method, and the m classification models are independently trained using the same data set, and m>n, m is an integer greater than 2, and n is an integer greater than 1.

[0009] As an optional implementation, obtaining a methylation dataset of a target object includes:

[0010] Acquire methylation data corresponding to a target methylation site of a target object, and determine the methylation data set according to the methylation data corresponding to the target methylation site;

[0011] The target methylation site is determined based on the degree of difference in methylation data of the same methylation site corresponding to different cancer staging values, or the correlation between the methylation data corresponding to the methylation site and different cancer staging values.

[0012] As an optional implementation, the staging prediction model further includes a standardization model, and the standardization model is used to standardize the methylation data in the methylation data set by means of de-meaning and variance normalization.

[0013] As an optional implementation, the staging prediction model further includes a dimensionality reduction model, and the dimensionality reduction model is used to perform dimensionality reduction processing on the methylation data in the methylation data set according to a PCA method.

[0014] As an optional implementation, the staging prediction model is determined in the following manner:

[0015] Using the data set to independently train the m classification models, respectively, to determine the classification accuracy of each of the m classification models;

[0016] According to the classification accuracy, k classification models are selected from the m classification models.

[0017] As an optional implementation manner, after selecting k classification models from m classification models according to the classification accuracy, the method further includes:

[0018] Using a cross-validation method, the n classification models are selected from the k classification models, wherein m≥k≥n; and / or,

[0019] The model parameters of the screened classification model are adjusted using the grid retrieval method.

[0020] As an optional implementation, the method of using a cross-validation method to select the n classification models from the k classification models includes:

[0021] Using the data set, cross-validating the k classification models respectively to obtain evaluation index values ​​of the k classification models;

[0022] According to the evaluation index values ​​of the k classification models, n classification models are selected from the k classification models.

[0023] As an optional implementation manner, after obtaining the data set, the method further includes:

[0024] The methylation data corresponding to the target methylation site is screened out from the methylation data contained in the data set, and the m classification models are independently trained using the screened methylation data.

[0025] As an optional implementation, the data set includes sample sets of different target objects, the sample sets include methylation data corresponding to each methylation site, and one sample set corresponds to one cancer stage value; the target methylation site is determined by:

[0026] According to the cancer staging values ​​corresponding to the sample sets, the data set is divided into a plurality of subsets, wherein each subset corresponds to a cancer staging value, and the sample sets included in different subsets correspond to different cancer staging values;

[0027] For each methylation site, a target methylation site is determined according to the methylation data in multiple subsets respectively corresponding to the methylation site.

[0028] As an optional implementation manner, determining the target methylation site according to the methylation data in multiple subsets respectively corresponding to the methylation site includes:

[0029] If the data set is divided into two subsets, then for each methylation site, the target methylation site is determined according to the difference between the methylation data corresponding to the methylation site in the two subsets; or,

[0030] If the data set is divided into two or more subsets, for each methylation site, a target methylation site is determined according to the correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset.

[0031] As an optional implementation, the difference degree of the methylation data corresponding to the methylation sites in the two subsets is determined by the following method:

[0032] Performing a rank sum test on the methylation data corresponding to the methylation sites in the two subsets respectively, and determining the degree of difference according to the rank sum test value; or,

[0033] The correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset is determined in the following manner:

[0034] A chi-square test is performed on the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset, and the correlation is determined according to the chi-square test value.

[0035] As an optional implementation, the multiple subsets include more than two subsets; and the dividing the data set into multiple subsets according to the cancer staging values ​​corresponding to the sample set further includes:

[0036] If, among the multiple divided subsets, there is a subset whose number of sample sets is lower than a threshold, the subset is merged into another subset according to the cancer stage value corresponding to the subset, and the cancer stage value corresponding to the merged another subset is determined, so as to use the merged subsets to determine the target methylation site and train the classification model; wherein the cancer stage value corresponding to the another subset is closest to the cancer stage value corresponding to the subset.

[0037] As an optional implementation, it also includes:

[0038] For the sample set of the target object, the methylation data corresponding to the methylation sites of the adjacent normal tissues in the sample set and / or the methylation data corresponding to the methylation sites containing missing values ​​are removed.

[0039] As an optional implementation, the stage prediction model is integrated according to the following method:

[0040] Using the outputs of n-1 classification models as the input of the nth classification model, and integrating the n classification models into the stage prediction model;

[0041] The input of the n-1 classification models is the data set, and the output of the nth classification model is the cancer stage value corresponding to the data set.

[0042] In a second aspect, an embodiment of the present disclosure further provides an electronic device, including a processor and a memory, wherein the memory is used to store a program executable by the processor, and the processor is used to read the program in the memory and perform the following steps:

[0043] Obtain a methylation dataset of a target object;

[0044] Inputting the methylation dataset into a staging prediction model, and outputting a cancer staging value of the target object;

[0045] Among them, the stage prediction model is obtained by integrating n classification models, the model parameters of the n classification models are optimized by using the grid retrieval method, the n classification models are screened from m classification models by using the cross-validation method, and the m classification models are independently trained using the same data set, and m>n, m is an integer greater than 2, and n is an integer greater than 1.

[0046] As an optional implementation, the processor is specifically configured to execute:

[0047] Acquire methylation data corresponding to a target methylation site of a target object, and determine the methylation data set according to the methylation data corresponding to the target methylation site;

[0048] The target methylation site is determined based on the degree of difference in methylation data of the same methylation site corresponding to different cancer staging values, or the correlation between the methylation data corresponding to the methylation site and different cancer staging values.

[0049] As an optional implementation, the staging prediction model further includes a standardization model, and the standardization model is used to standardize the methylation data in the methylation data set by means of de-meaning and variance normalization.

[0050] As an optional implementation, the staging prediction model further includes a dimensionality reduction model, and the dimensionality reduction model is used to perform dimensionality reduction processing on the methylation data in the methylation data set according to a PCA method.

[0051] As an optional implementation, the processor is specifically configured to determine the staging prediction model in the following manner:

[0052] Using the data set to independently train the m classification models, respectively, to determine the classification accuracy of each of the m classification models;

[0053] According to the classification accuracy, k classification models are selected from the m classification models.

[0054] As an optional implementation manner, after selecting k classification models from m classification models according to the classification accuracy, the processor is further configured to execute:

[0055] Using a cross-validation method, the n classification models are selected from the k classification models, wherein m≥k≥n; and / or,

[0056] The model parameters of the screened classification model are adjusted using the grid retrieval method.

[0057] As an optional implementation, the processor is specifically configured to execute:

[0058] Using the data set, cross-validating the k classification models respectively to obtain evaluation index values ​​of the k classification models;

[0059] According to the evaluation index values ​​of the k classification models, n classification models are selected from the k classification models.

[0060] As an optional implementation, after acquiring the data set, the processor is further configured to execute:

[0061] The methylation data corresponding to the target methylation site is screened out from the methylation data contained in the data set, and the m classification models are independently trained using the screened methylation data.

[0062] As an optional implementation, the data set includes sample sets of different target objects, the sample sets include methylation data corresponding to each methylation site, and one sample set corresponds to one cancer stage value; the processor is specifically configured to determine the target methylation site in the following manner:

[0063] According to the cancer staging values ​​corresponding to the sample sets, the data set is divided into a plurality of subsets, wherein each subset corresponds to a cancer staging value, and the sample sets included in different subsets correspond to different cancer staging values;

[0064] For each methylation site, a target methylation site is determined according to the methylation data in multiple subsets respectively corresponding to the methylation site.

[0065] As an optional implementation, the processor is specifically configured to execute:

[0066] If the data set is divided into two subsets, then for each methylation site, the target methylation site is determined according to the difference between the methylation data corresponding to the methylation site in the two subsets; or,

[0067] If the data set is divided into two or more subsets, for each methylation site, a target methylation site is determined according to the correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset.

[0068] As an optional implementation, the processor is specifically configured to determine the degree of difference between the methylation data in two subsets respectively corresponding to the methylation sites in the following manner:

[0069] Performing a rank sum test on the methylation data corresponding to the methylation sites in the two subsets respectively, and determining the degree of difference according to the rank sum test value; or,

[0070] The correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset is determined in the following manner:

[0071] A chi-square test is performed on the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset, and the correlation is determined according to the chi-square test value.

[0072] As an optional implementation, the multiple subsets include more than two subsets; the data set is divided into multiple subsets according to the cancer staging values ​​corresponding to the sample set, and the processor is further configured to execute:

[0073] If, among the multiple divided subsets, there is a subset whose number of sample sets is lower than a threshold, the subset is merged into another subset according to the cancer stage value corresponding to the subset, and the cancer stage value corresponding to the merged another subset is determined, so as to use the merged subsets to determine the target methylation site and train the classification model; wherein the cancer stage value corresponding to the another subset is closest to the cancer stage value corresponding to the subset.

[0074] As an optional implementation, the processor is further configured to execute:

[0075] For the sample set of the target object, the methylation data corresponding to the methylation sites of the adjacent normal tissues in the sample set and / or the methylation data corresponding to the methylation sites containing missing values ​​are removed.

[0076] As an optional implementation, the stage prediction model is integrated according to the following method:

[0077] Using the outputs of n-1 classification models as the input of the nth classification model, and integrating the n classification models into the stage prediction model;

[0078] The input of the n-1 classification models is the data set, and the output of the nth classification model is the cancer stage value corresponding to the data set.

[0079] In a third aspect, an embodiment of the present disclosure further provides a computer storage medium on which a computer program is stored, and when the program is executed by a processor, it is used to implement the steps of the method described in the first aspect above.

[0080] These and other aspects of the present disclosure will become more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0082] Figure 1 A specific implementation flow chart of a cancer staging method provided by an embodiment of the present disclosure;

[0083] Figure 2 A schematic diagram of a data format after a data set is preprocessed according to an embodiment of the present disclosure;

[0084] Figure 3 A schematic diagram of a differential analysis p-value of methylation data corresponding to each methylation site provided in an embodiment of the present disclosure;

[0085] Figure 4 A distribution diagram of methylation data before normalization provided in an embodiment of the present disclosure;

[0086] Figure 5 A distribution diagram of methylation data after standardization and dimension reduction provided in an embodiment of the present disclosure;

[0087] Figures 6A to 6E A schematic diagram of a cross-validation classification model evaluation provided in an embodiment of the present disclosure;

[0088] Figure 7 A flowchart of multi-model integration provided for an embodiment of the present disclosure;

[0089] Figure 8 A ROC_AUC indicator evaluation curve diagram of a staging prediction model provided in an embodiment of the present disclosure;

[0090] Fig. 9 A result diagram of cancer staging prediction for liver cancer patients using a staging prediction model provided in an embodiment of the present disclosure;

[0091] Fig.10 A specific implementation flow chart of a cancer staging method provided by an embodiment of the present disclosure;

[0092] Fig.11 A schematic diagram of a cancer staging device provided in an embodiment of the present disclosure;

[0093] Fig.12 A schematic diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0094] In order to make the purpose, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0095] In the embodiments of the present disclosure, the term "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0096] The term "methylation" in the embodiments of the present disclosure is catalyzed by enzymes in biological systems. Methylation involves heavy metal modification, regulation of gene expression, regulation of protein function, and ribonucleic acid (RNA) processing. Gene methylation is the most important form of epigenetic inheritance and can affect the occurrence and development of tumors.

[0097] The term "cancer staging" in the embodiments of the present disclosure is a method for determining the extent of cancer development and spread. Cancer staging affects the patient's treatment plan and the determination of the prognosis of cancer.

[0098] The term “PCA (Principal Component Analysis)” in the embodiments of the present disclosure is often used for dimensionality reduction of high-dimensional data and can be used to extract the main features of the data.

[0099] The term "grid retrieval" in the embodiments of the present disclosure is a model hyperparameter optimization technology, which is often used to optimize three or fewer hyperparameters and is essentially an exhaustive method.

[0100] The term "model integration" in the disclosed embodiments refers to combining multiple weakly supervised models in order to obtain a better and more comprehensive strongly supervised model. The underlying idea of ​​the integrated model is that even if a weak classifier makes an incorrect prediction, other weak classifiers can correct the error.

[0101] The application scenarios described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. It is known to those skilled in the art that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems. In the description of the present disclosure, unless otherwise specified, the meaning of "multiple" is two or more.

[0102] Cancer staging is a method to determine the extent of cancer development and spread. Accurate cancer staging is conducive to formulating the most reasonable treatment plan for cancer patients and effectively judging the prognosis of cancer. In blood tumors, lymphoma and Hodgkin lymphoma both use the Ann Arbor staging system; in solid tumors, TNM (Tumor Node Metastasis) staging is the most common staging system, but there are many differences due to different tumor types. Commonly used tumor types include: breast cancer, colon cancer, kidney cancer, laryngeal cancer, liver cancer, lung cancer, prostate cancer, skin cancer, bladder cancer, etc., which can be divided into I, II, III, and IV stages according to the degree of tumor damage. Since the existing technology mostly uses clinical and pathological examinations to stage cancer, the disadvantage is that when only clinical examinations (blood tests, radiological examinations, endoscopic examinations, etc.) are used, the implementation method is indirect observation, and the information for cancer staging judgment is limited, which has limitations. Although pathological examination can determine the stage more accurately, it requires surgical operation to obtain pathological sections. Not all tumors require surgical treatment, and some patients may have undergone chemotherapy and radiotherapy before pathological examination, which will underestimate the true stage of the tumor to a certain extent. Therefore, the method of using pathological examination still has limitations. Secondly, different cancers have different staging systems, and some cancers do not have suitable staging methods. The staging is highly subjective, and conventional staging methods are not universal. In addition to clinical systems, there are currently methods for cancer staging based on methylation data, but these methods usually only consider a few methylation sites, and the classification model method is relatively simple, and fails to effectively utilize most of the methylation data of the genome.

[0103] The cancer staging method provided in this embodiment is used to predict the cancer stage corresponding to the input methylation data set by using a staging prediction model obtained by integrating multiple classification models. The staging prediction model in this embodiment is independently trained by m classification models, and n classification models are screened out by a cross-validation method, and the model parameters of these n classification models are optimized by a grid retrieval method, which can effectively capture the complex relationship between methylation data and cancer stage, and improve the accuracy of cancer stage prediction; compared with the current clinical staging method, cancer stage prediction using methylation site information is applicable to most cancer types, can effectively distinguish the cancer stage types of different patients, and improve the universality of cancer stage prediction.

[0104] like Figure 1 As shown, the specific implementation process of a cancer staging method provided in this embodiment is as follows:

[0105] Step 100, obtaining a methylation dataset of a target object;

[0106] The target object in this embodiment includes but is not limited to human tumor tissue.

[0107] In implementation, the methylation data set in this embodiment includes methylation data corresponding to multiple methylation sites, wherein one methylation site of one sample corresponds to one methylation data. The methylation data in this embodiment includes methylation level values.

[0108] In some embodiments, the methylation dataset of the target object is determined by:

[0109] Acquire methylation data corresponding to a target methylation site of a target object, and determine the methylation data set according to the methylation data corresponding to the target methylation site;

[0110] The target methylation site is determined based on the degree of difference in methylation data of the same methylation site corresponding to different cancer staging values, or the correlation between the methylation data corresponding to the methylation site and different cancer staging values.

[0111] In implementation, the methylation data set of the target object in this embodiment includes the methylation data corresponding to the target methylation site, and the target methylation site is screened out from each methylation site. Since the methylation data corresponding to the target methylation site can more accurately characterize the characteristics of cancer staging, the methylation data set formed by using the methylation data corresponding to the target methylation site can more accurately predict the cancer staging and improve the accuracy of the prediction.

[0112] It should be noted that the number of target methylation sites is less than or equal to the number of each methylation site contained in the target object itself.

[0113] Step 101, inputting the methylation dataset into a staging prediction model, and outputting the cancer staging value of the target object;

[0114] Among them, the stage prediction model is obtained by integrating n classification models, the model parameters of the n classification models are optimized by using the grid retrieval method, the n classification models are screened from m classification models by using the cross-validation method, and the m classification models are independently trained using the same data set, and m>n, m is an integer greater than 2, and n is an integer greater than 1.

[0115] In implementation, the staging prediction model in this embodiment is used to predict cancer staging for the input methylation data set. On the one hand, since the staging prediction model of this embodiment is integrated by multiple classification models, and model integration refers to combining multiple weakly supervised models in order to obtain a better and more comprehensive strong supervision model, the underlying idea of ​​the integrated model is that even if a certain weak classifier obtains an incorrect prediction, other weak classifiers can correct the error. Therefore, the accuracy of the staging prediction model prediction is improved by model integration. On the other hand, the n classification models included in the staging prediction model in this embodiment are screened from m classification models using a cross-validation method, and cross-validation can evaluate whether the m classification models perform well on the entire data set, and is used to evaluate the stability and generalization ability of the classification model. For example, when one of the m classification models fluctuates greatly on the AUC index of all folds of cross-validation, the classification model will be excluded, that is, it will not be selected as one of the n classification models. Finally, the model parameters of the n classification models finally selected in this embodiment are optimized using the grid retrieval method, and network retrieval is a model hyperparameter optimization technology, which is often used to optimize three or fewer hyperparameters and is essentially an exhaustive method. The purpose of adjusting the model parameters of the n classification models using grid retrieval is to select the model parameters with the best classification model performance based on the n classification models that have been selected. This embodiment optimizes the accuracy, stability and prediction ability of the stage prediction model for new data sets through the coordinated execution of independent training, cross-validation and grid retrieval parameter adjustment of multiple classification models.

[0116] In some embodiments, the staging prediction model further includes a standardization model, which is used to standardize the methylation data in the methylation dataset by removing the mean and normalizing the variance.

[0117] In practice, since the distribution of methylation data of multiple target methylation sites is skewed, in order to make the methylation rates of the methylation data of multiple target methylation sites uniform in scale, the methylation data of each target methylation site can be standardized so that the methylation data of each target methylation site after processing is as close to the normal distribution as possible. This makes the subsequent prediction based on the methylation data of each target methylation site more accurate.

[0118] In some embodiments, the staging prediction model further includes a dimensionality reduction model, and the dimensionality reduction model is used to perform dimensionality reduction processing on the methylation data in the methylation dataset according to a PCA method.

[0119] In practice, PCA transforms the original data into a set of linearly independent representations of each dimension through linear transformation, which can be used to extract the main characteristic components of the data and is often used for dimensionality reduction of high-dimensional data. PCA extracts the main features of the methylation data set formed by each target methylation data, thereby obtaining a methylation data set containing the main features, so that when subsequent predictions are made based on the methylation data set containing the main features, the amount of data that needs to be calculated is reduced, and the prediction speed and efficiency of the algorithm are improved.

[0120] Optionally, the stage prediction model in this embodiment includes, but is not limited to, normalizing and reducing the methylation data set, and using the normalized and reduced methylation data set for prediction. This embodiment does not impose too many restrictions on the order of normalization and dimensionality reduction.

[0121] In some embodiments, this embodiment determines the staging prediction model in the following manner:

[0122] Step 1: Use the data set to independently train the m classification models to determine the classification accuracy of each of the m classification models;

[0123] In implementation, the data set includes sample sets of different target objects, and the sample sets include methylation data corresponding to each methylation site; one sample set corresponds to one cancer staging value; one target object corresponds to one sample set, and one sample set includes methylation data corresponding to multiple methylation sites.

[0124] In implementation, during the training of the classification model, a training set in the data set is used for training, and the training set includes a sample set.

[0125] Optionally, after obtaining the data set and before using the data set to train the classification model, the methylation data in the data set may be preprocessed. The preprocessing method includes but is not limited to any one or a combination of any of the following methods:

[0126] Method 1a: After obtaining the sample set, the methylation data corresponding to the methylation sites of the normal tissue adjacent to the cancer in the sample set may be removed for the sample set of the target object;

[0127] Mode 1b: After obtaining the sample set, the methylation data corresponding to the methylation sites containing missing values ​​(NA) in the sample set may be removed for the sample set of the target object.

[0128] During implementation, a sample set can be obtained from the public database TCGA, each sample set corresponds to a cancer staging value, and the methylation data corresponding to the methylation sites of the adjacent normal tissue and the methylation data corresponding to the methylation sites of the tumor tissue containing missing values ​​are removed.

[0129] In some embodiments, after obtaining the data set, the present embodiment may further screen the training set in the data set again to screen out the methylation data corresponding to the target methylation site that characterizes the cancer staging characteristics, thereby improving the prediction accuracy. The specific method is as follows:

[0130] Mode 1c: Filter out the methylation data corresponding to the target methylation site from the methylation data included in the training set in the data set, and use the filtered methylation data to independently train the m classification models. The filtered methylation data refers to the methylation data corresponding to the filtered target methylation site.

[0131] In implementation, after obtaining a sample set and removing the methylation data corresponding to the methylation sites of adjacent normal tissues and the methylation data corresponding to the methylation sites containing missing values, the methylation data corresponding to the target methylation sites can be screened out in the removed sample set, thereby forming a data set using the screened sample set.

[0132] In some embodiments, the target methylation site is determined according to the training set in the data set, and the target methylation site can be determined by the following process:

[0133] Process c1) dividing the data set into multiple subsets according to the cancer staging values ​​corresponding to the sample sets, wherein each subset corresponds to one cancer staging value, and the sample sets included in different subsets correspond to different cancer staging values;

[0134] In implementation, the data set can be divided into two subsets or more than two subsets. For example, one subset corresponds to a cancer staging value, and the sample set of stage I is divided into one subset, the sample set of stage II is divided into one subset, the sample set of stage III is divided into one subset, and the sample set of stage IV is divided into one subset, so that the data set is finally divided into 4 subsets. The sample sets of stages I and II can also be divided into one subset, and the sample sets of stages III and IV can be divided into one subset, that is, the data set is divided into two subsets, and the cancer staging value corresponding to a subset of the sample set containing stages I and II is set to the early stage, and the cancer staging value corresponding to a subset of the sample set containing stages III and IV is set to the late stage. This embodiment does not impose too many restrictions on the division of the data set, and it can be adjusted according to actual conditions.

[0135] Optionally, after the data set is divided into multiple subsets according to the cancer staging values ​​corresponding to the sample set, if the multiple subsets include more than two subsets, the subsets may be merged. The specific steps are as follows:

[0136] If, among the multiple divided subsets, there is a subset whose number of sample sets is lower than a threshold, the subset is merged into another subset according to the cancer stage value corresponding to the subset, and the cancer stage value corresponding to the merged another subset is determined, so as to use the merged subsets to determine the target methylation site and train the classification model; wherein the cancer stage value corresponding to the another subset is closest to the cancer stage value corresponding to the subset.

[0137] In implementation, when there are more than two subsets, if there is an imbalance in the amount of methylation data between different subsets, different subsets can be merged. For example, the data set includes sample sets of different liver cancer patients and corresponding cancer stage values, where the number of stage I cases (the number of sample sets) is 212, the number of stage II cases is 107, and the number of stages III and IV is 150 in total. The methylation sites with missing methylation data are removed, and the methylation data of normal samples are removed. The sample sets are grouped according to the cancer stage value, that is, the data set is split. Since the number of cases in stages III and IV is scattered, and the number of cases in some periods is small, there is an imbalance in the sample set, so the cases in stages III and IV are merged into the late stage of cancer, and the number of cases in stages I and II are merged into the early stage of cancer. So far, the early stage of cancer contains 265 cases, and the late stage of cancer contains 115 cases.

[0138] Optionally, after dividing the data set into multiple subsets, the data set can be formatted to facilitate the next process. For example, the data set can be transposed so that the first column contains cancer stage values, and the second column and subsequent columns contain methylation data corresponding to the methylation sites, such as Figure 2 As shown, this embodiment provides a data format after a data set is preprocessed.

[0139] After the data set is preprocessed, the preprocessed data set is split into a training set and a test set at a ratio of 8:2. In the specific implementation, the data set is first divided into stages, that is, the data set is divided into multiple subsets, and then the data set is split into a training set and a test set. Then, the split training set includes multiple subsets, and the test set includes multiple subsets.

[0140] Process c2) For each methylation site, a target methylation site is determined according to the methylation data corresponding to the methylation site in multiple subsets.

[0141] According to the number of subsets of the training set divided into, this embodiment provides different methods for determining the target methylation site, which are specifically as follows:

[0142] Method c1: The training set includes two subsets, and the target methylation sites are determined by the following method:

[0143] For each methylation site, the target methylation site is determined according to the difference degree of the methylation data corresponding to the methylation site in the two subsets.

[0144] In implementation, the difference degree of the methylation data corresponding to the methylation sites in the two subsets is determined by the following method:

[0145] A rank sum test is performed on the methylation data corresponding to the methylation sites in the two subsets respectively, and the degree of difference is determined according to the rank sum test value.

[0146] In the implementation, according to the cancer stage, the training set includes subset 1 corresponding to the early stage of cancer and subset 2 corresponding to the late stage. The methylation data corresponding to the same methylation site in the two subsets are subjected to a rank sum test to obtain the rank sum test value p. The formula is as follows:

[0147] P n =w(G1,G2) Formula 1;

[0148] In formula 1, P n is the rank sum test value (significance test value) obtained after the rank sum test of the nth methylation site, w is the rank sum test function, G1 represents the methylation data of the nth methylation site in subset 1, and G2 represents the methylation data of the nth methylation site in subset 2.

[0149] Then, sort by p-value from small to large, and select the top 1,000 methylation sites with the highest degree of difference (i.e., the smallest p-value) as target methylation sites. In theory, the methylation data corresponding to these 1,000 methylation sites can more accurately distinguish target objects under different cancer stages.

[0150] For example, the preprocessed data set is classified into early cancer (marked as 0) and late cancer (marked as 1), and two subsets are obtained, one subset corresponds to early cancer and the other subset corresponds to late cancer. The methylation data at each methylation site are differentially analyzed using the rank sum test, and the p value is calculated. The p values ​​are sorted from small to large, and the top 1000 methylation sites are selected as target methylation sites. Figure 3 As shown, this embodiment provides a schematic diagram of the differential analysis p-value of the methylation data corresponding to each methylation site, wherein the leftmost side represents the index value of the methylation site, the column corresponding to cg represents the identifier of the methylation site, and pvalue represents the rank sum test p-value calculated from the methylation data corresponding to the methylation site.

[0151] When analyzing methylation data, the rank sum test can be used to detect differential methylation sites. Since the number of sample sets under different cancer stages is usually different and there is no obvious pattern in the distribution of methylation rates, the rank sum test can be used to perform significance tests under conditions where the data does not conform to the normal distribution and the number of sample sets is different. The methylation data corresponding to the top 1000 most significant target methylation sites are obtained.

[0152] Method c2, the training set includes more than two subsets, and the target methylation sites are determined by the following method:

[0153] For each methylation site, a target methylation site is determined based on the correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset.

[0154] In implementation, the correlation between the methylation data corresponding to the methylation sites in each subset and the cancer stage value corresponding to the subset is determined in the following manner:

[0155] A chi-square test is performed on the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset, and the correlation is determined according to the chi-square test value.

[0156] Among them, the chi-square test is the degree of deviation between the actual observed value of the statistical sample and the theoretical inference value. The degree of deviation between the actual observed value and the theoretical inference value determines the size of the chi-square value.

[0157] In practice, if the training set includes three or more subsets, each subset corresponds to a cancer stage, and the chi-square test can be used to calculate the correlation between each methylation site and the cancer stage value, that is, the chi-square statistic, and select the 1000 methylation sites with the highest chi-square statistic as the target methylation sites. The formula for calculating the chi-square test is as follows:

[0158] S n =chi2(X n ,y) Formula 2;

[0159] In formula 2, S n represents the chi-square test value obtained after the chi-square test of the nth methylation site, chi2 represents the chi-square test function, X n represents the methylation data corresponding to the nth methylation site in each subset, and y represents the cancer stage value corresponding to each subset.

[0160] When the training set is divided into three subsets, the value of y is 0, 1, and 2; when the training set is divided into four subsets, the value of y is 0, 1, 2, and 3.

[0161] Optionally, after the target methylation sites are determined as described above, the training set can also be preprocessed by standardization and dimensionality reduction. For example, after 1000 target methylation sites are determined, the methylation data corresponding to the 1000 methylation sites are subjected to data standardization. The standardization process uses mean removal and variance normalization, and the formula is as follows:

[0162]

[0163] In formula 3, Represents the methylation data corresponding to the target methylation site after normalization, M n represents the methylation data corresponding to the target methylation site, μ represents the mean of each methylation data corresponding to the target methylation site in each subset, and σ represents the standard deviation of the methylation data corresponding to the target methylation site in each subset.

[0164] For example, statistics are performed on the methylation data corresponding to 1000 target methylation sites to obtain a distribution diagram such as Figure 4 As shown, the present embodiment provides a distribution diagram of methylation data before normalization. The distribution of methylation data corresponding to each target methylation site in the diagram is skewed, and skewed distribution is not conducive to model training. Therefore, the skewed data is standardized by removing the mean and normalizing the variance to obtain methylation data that is as close to a normal distribution as possible to train the classification model.

[0165] Optionally, after the methylation data corresponding to the 1000 methylation sites are subjected to data standardization, the PCA dimensionality reduction method can be used to reduce the methylation data corresponding to the 1000 methylation sites to 10 feature data. The formula is as follows:

[0166] X * =PCA(n).fit_transform(X) Formula 4;

[0167] In formula 4, X * represents the methylation data after dimension reduction processing of X, X * It can be understood as a matrix. PCA(n) represents the construction of a PCA function with a dimension reduction result of n. n represents the dimension of the methylation data after dimension reduction. For example, n=10. The function fits and transforms X. X represents the methylation data before dimension reduction.

[0168] In the implementation, since too many features will affect the model's search for internal rules, and the methylation data features in the training set in this embodiment are extremely complex, the PCA dimensionality reduction method can be used to set the dimension to 10, and reduce the high-dimensional 1000 feature vectors (methylation data corresponding to 1000 methylation sites) to low-dimensional 10 feature vectors (methylation data corresponding to 10 methylation sites). While speeding up the classification model training speed, the characteristics of the metadata can also be guaranteed, such as Figure 5 As shown, this embodiment provides a distribution diagram of methylation data after standardization and dimensionality reduction processing. The distribution of the processed methylation data is close to a normal distribution, which is more conducive to the training of a classification model.

[0169] Optionally, to meet the needs of specific data set analysis, the order of rank sum test / chi-square test, standardization and PCA dimensionality reduction can be freely swapped. For example, when processing larger data sets, such as whole genome methylation sequencing data, since the number of methylation sites in the genome is generally tens of thousands, PCA dimensionality reduction can be prioritized to reduce the dimension of the data and reduce the time of rank sum test / chi-square test and standardization. When the number of methylation sites is larger, the time saved is more obvious. In addition, if the targeted methylation sequencing data containing only dozens of genes is processed, standardization can be performed first, and then rank sum test / chi-square test and PCA dimensionality reduction can be performed. After standardization, the data distribution is more uniform. When performing significance tests, since it is not affected by the numerical size of the data itself, differential methylation sites can be detected more easily, improving the sensitivity of rank sum test / chi-square test, and similarly, the effect of PCA dimensionality reduction can also be improved.

[0170] After the training set is processed by the above process, the processed training set is input into the traversal fitting of the classification model algorithm, and the training set is traversed and input into m classification models, and the model performance of each classification model is calculated. Optionally, the m classification models in this embodiment include but are not limited to logistic regression (LogisticRegression), multi-layer perceptron (MLPClassifier), K nearest neighbor (KNeighborsClassifier), support vector machine (SVC), Gaussian process (GaussianProcessClassifier), decision tree (DecisionTreeClassifier), Gaussian naive Bayes (GaussianNB), random forest (RandomForestClassifier), discriminant analysis (DiscriminantAnalysis), multi-layer perceptron (MLPClassifier) ​​and other models. The classification accuracy of each classification model is calculated as an indicator for judging the performance of the model.

[0171] Step 2: According to the classification accuracy, k classification models are selected from the m classification models.

[0172] In implementation, the top k classification models are selected from the m classification models in descending order of classification accuracy for subsequent cross-validation and / or grid retrieval (adjusting model parameters).

[0173] Optionally, after selecting k classification models from m classification models according to the classification accuracy, any one or more of the following may be performed:

[0174] a) using a cross-validation method to select the n classification models from the k classification models, where m≥k≥n;

[0175] b) Using the grid retrieval method, the model parameters of the screened classification model are adjusted.

[0176] The above cross validation and grid retrieval can be performed simultaneously, but the order of execution can be customized according to actual needs. When the data set is large, you can first perform cross validation to screen out n classification models, and then use the grid retrieval method to adjust the model parameters of the screened n classification models. When the data set is small, you can first use the grid retrieval method to adjust the model parameters of the screened k classification models, and then use the cross validation method to screen out the n classification models from the k classification models.

[0177] In implementation, the cross-validation method is used to select the n classification models from the k classification models. The cross-validation formula is as follows:

[0178] C k =T(X train ,y train ) Formula 5a;

[0179] Score k =C k (X test ,y test ) Formula 5b;

[0180] Among them, X train ,y train Respectively represent the methylation data in the training set and the cancer stage value corresponding to the training set; X test ,y test represents the methylation data in the test set and the cancer stage value corresponding to the test set respectively; wherein the training set and the test set are the data set segmentation conditions under the k-th fold of the data set, T represents the multiple classification models mentioned in this embodiment, and C k Indicates the classification model generated by fitting under the k-th fold, Score k C is the k-th fold classification modelk According to X test ,y test Calculate the model AUC (evaluation index value).

[0181] Optionally, perform cross-validation and screening as follows:

[0182] Step 3a) using the data set, cross-validating the k classification models respectively to obtain evaluation index values ​​of the k classification models;

[0183] During implementation, the model accuracy of the classification model may be used as an evaluation index value.

[0184] Step 3b) select n classification models from the k classification models according to the evaluation index values ​​of the k classification models respectively.

[0185] In implementation, the above Score can be used k The AUC averages are used to rank the classification models from high to low, and the top n classification models are selected.

[0186] For example, to prevent overfitting, a 5-fold cross validation with K=5 is performed on the data set, ROC_AUC is selected as the validation evaluation method, and a plot is made. The top five classification models with better cross validation are selected, including logistic regression, linear discriminant analysis, extreme random tree, Gaussian process, and K nearest neighbor classification models. The fitting method is better, such as Figures 6A to 6E As shown, this embodiment provides a schematic diagram of a cross-validation classification model evaluation, wherein, Figures 6A to 6E Represents the ROC_AUC graphs of the logistic regression, linear discriminant analysis, extreme random tree, Gaussian process, and K nearest neighbor classification models respectively.

[0187] , and also performs the following parameter optimization process:

[0188] In some embodiments, the model parameters of the screened classification models are adjusted using a grid retrieval method to obtain an adjusted classification model. Taking cross-validation followed by grid retrieval as an example, after the n classification models are screened out from k classification models using a cross-validation method, the model parameters of the classification models among the n classification models are adjusted using a grid retrieval method, and the final staging prediction model is determined based on the adjusted classification model.

[0189] In the implementation, after selecting n classification models with good cross-validation, the model parameters of the classification models in the n classification models are grid retrieved and adjusted. The model parameters involved in the n classification models are arranged, combined and configured, and the optimal combination of model parameter configuration is obtained according to the calculation results of the model accuracy.

[0190] For example, the grid retrieval method is used to adjust the parameters of the above-mentioned classification models of logistic regression, linear discriminant analysis, extreme random tree, and Gaussian process, and the ROC_AUC indicator is used for evaluation. The four classification models with adjusted parameters, logistic regression, linear discriminant analysis, Gaussian process, and extreme random tree, are subsequently integrated.

[0191] In some embodiments, the staging prediction model is integrated according to the following method:

[0192] Using the outputs of n-1 classification models as the input of the nth classification model, and integrating the n classification models into the stage prediction model;

[0193] The input of the n-1 classification models is the training set, and the output of the nth classification model is the cancer stage value corresponding to the training set.

[0194] In the implementation, the n classification models with adjusted parameters are integrated to generate the final staging prediction model. For example, when n=4, the features output by the three classification models are used as the input data of the fourth classification model for fitting to obtain the final staging prediction model. Figure 7 As shown, this embodiment provides a flowchart of multi-model integration, inputs methylation data in the training set into multiple classification models for fitting training, evaluates multiple classification models according to model evaluation indicators, and selects n classification models for model integration. The model in the figure refers to the classification model.

[0195] The formula for model integration is as follows:

[0196] C combine =C4(S C1 ,S C2 ,S C3 ) Formula 6;

[0197] In formula 6, C combine represents the integrated stage prediction model, S C1 ,S C2 ,S C3 Represent the predicted probability scores generated by classification model C1, classification model C2 and classification model C3 respectively. C1 ,S C2 ,S C3 The output data is input into the fourth classification model C4 for model construction to obtain the final integrated model.

[0198] For example, four classification models with good parameter adjustment (logistic regression, linear discriminant analysis, Gaussian process, extreme random tree) are selected, and the stacking method in mlxtend is used for model integration. The output data of the first three classification models are used as the input data of the last classification model for fitting. At the same time, the ROC_AUC indicator is used for evaluation to generate a staging prediction model. The ROC_AUC indicator evaluation curve of the staging prediction model is as follows: Figure 8 shown.

[0199] After the staging prediction model is integrated using the above method, in the use stage, the methylation data of patients with unknown liver cancer staging are first obtained, and then 1000 target methylation sites corresponding to the staging prediction model are selected. Finally, the methylation data corresponding to the 1000 target methylation sites are standardized and dimensionally reduced, and the standardized and dimensionally reduced methylation data are input into the staging prediction model for staging prediction, and the corresponding prediction results are output, such as Fig. 9 As shown, this embodiment provides a result diagram of cancer staging prediction for liver cancer patients using a staging prediction model. The diagram includes the predicted staging of each liver cancer patient and indicates the specific staging probability.

[0200] This embodiment uses the traversal fitting classification algorithm (using the training set to independently train m classification models), cross validation and grid retrieval parameter adjustment in coordination, wherein the traversal fitting can preliminarily select the candidate better classification algorithm from all the classification algorithms, and the cross validation can evaluate whether the aforementioned candidate classification algorithm performs well on the entire data set, which is used to evaluate the stability and generalization ability of the classification model. If the candidate classification model selected by the traversal fitting fluctuates greatly in the AUC index of all cross validation folds, the classification algorithm used by the classification model will be excluded. The model parameter adjustment is to select the parameters with the best model performance based on the n selected classification models. After the traversal fitting classification algorithm, cross validation and grid retrieval parameter adjustment, the classification model has the best accuracy, stability and prediction ability for the new data set. Model integration is used to further integrate the aforementioned better n models to obtain the final stage prediction model, and the performance of the final stage prediction model is higher than any one of the n models.

[0201] Optionally, if the amount of data is not large, such as when processing targeted methylation sequencing data containing only dozens of genes, the idea of ​​traversing and fitting the classification model, grid retrieval, and then cross-validation can be used to build a model. This operation can evaluate all classification algorithms and their parameters. Theoretically, the conditions considered are the most sufficient and the most comprehensive when screening models. However, because the computational complexity of considering all classification algorithms and their parameters increases exponentially, it consumes the longest time. If the amount of data is large, such as when processing whole genome methylation sequencing data, the idea of ​​traversing and fitting the classification model first, then cross-validation, and finally grid retrieval can be used to build a model. Traversing and fitting the classification model can preliminarily select the candidate better classification algorithm from all classification algorithms. Cross-validation is used to evaluate the stability and generalization ability of the classification model. Grid retrieval (i.e., model parameter adjustment) is to select the parameters with the best model performance based on the n classification models that have been selected. After traversing and fitting the classification algorithm, cross-validation, and grid retrieval parameter adjustment, the classification model has the best accuracy, stability, and prediction ability for new data sets.

[0202] The idea of ​​model integration in the present embodiment is: the output data of the first three classification models are used as the input data of the last classification model for fitting, and the final stage prediction model is obtained. Specifically, the methylation data after standardization and dimensionality reduction are respectively input into the three classification models, and the prediction results of the prediction probability of each class of the three classification models output are used as the original feature data, and are input into the fourth classification model again for modeling. The method of modeling using the features generated by the first three classifiers as the input data of the last classifier, due to the use of the specific numerical values ​​of the probability of each classifier under different classifications, and the probability numerical value is used as the original data set for model construction evaluation, compared with the method of voting weighted integration, the different classification models are more perfect and integrated, and the weight of each classifier in the prediction result can be objectively considered, and the accuracy of the final model is theoretically improved. The method of voting weighted integration only makes a final judgment from the statistics of the proportion of the classification results of different classifiers. When the weights of different classifiers are not considered, the influence of the poorly effective classifier on the final result is aggravated, and some noise is introduced. When the weights of different classifiers are considered small, the setting of the weight size is more subjective.

[0203] like Fig.10 As shown, the specific implementation process of a cancer staging method provided in this embodiment is as follows:

[0204] Step 1000: Obtain a data set, remove the methylation data corresponding to the methylation sites of normal tissue adjacent to the cancer, and the methylation data corresponding to the methylation sites containing missing values;

[0205] Step 1001: Divide the data set into multiple subsets according to the cancer staging value, each subset corresponding to a cancer staging value;

[0206] Step 1002: for each methylation site, determine the target methylation site according to the methylation data corresponding to the methylation site in multiple subsets of the training set;

[0207] Step 1003, performing standardization and dimensionality reduction processing on the methylation data corresponding to the target methylation site in the training set to obtain a processed training set;

[0208] Step 1004: input the processed training sets into m classification models respectively, train the m classification models independently, and select k classification models according to the classification accuracy of each of the m classification models;

[0209] Step 1005: Using a cross-validation method, select the n classification models from the k classification models, and using a grid retrieval method, adjust the model parameters of the classification models in the n classification models to obtain the adjusted n classification models;

[0210] Step 1006: using the outputs of n-1 classification models as the inputs of the nth classification model, and integrating the n classification models into the stage prediction model;

[0211] Step 1007: Acquire a methylation dataset of the target object, input the methylation dataset into a staging prediction model, and output the cancer staging value of the target object.

[0212] The cancer staging method provided in this embodiment obtains the methylation data of cancer patients and the corresponding cancer staging values ​​to form a data set, preprocesses the data set and divides the subsets; then splits the data set into a training set and a test set at a ratio of 8:2, performs differential methylation analysis on the training set, selects multiple target methylation sites with the highest degree of difference, and uses the PCA method to reduce the methylation data corresponding to the multiple target methylation sites; uses multiple classification algorithms to traverse and fit the training set, uses the fitted classification model for the test set test, selects n classification models with better test results for grid retrieval and parameter adjustment, and integrates the classification models after parameter adjustment to obtain the final staging prediction model; and uses the final staging prediction model to predict the staging of the patient's cancer. This embodiment uses methylation site information to predict cancer staging. Compared with the current clinical staging method, it is applicable to most cancer types and can effectively distinguish the cancer staging types of different patients. The classification algorithm combined with differential methylation site screening, feature dimensionality reduction and multi-model integration can effectively capture the complex relationship between methylation data and cancer staging, and improve the accuracy of cancer staging prediction.

[0213] Based on the same inventive concept, the embodiment of the present disclosure also provides a cancer staging device. Since the device is the device in the method in the embodiment of the present disclosure, and the principle of solving the problem by the device is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0214] like Fig.11 As shown, the device comprises:

[0215] A data acquisition module 1100 is used to acquire a methylation dataset of a target object;

[0216] The staging prediction module 1101 is used to input the methylation data set into the staging prediction model and output the cancer staging value of the target object; wherein the staging prediction model is obtained by integrating n classification models, the model parameters of the n classification models are optimized by using a grid retrieval method, the n classification models are screened from m classification models by using a cross-validation method, the m classification models are independently trained using the same data set, and m>n, m is an integer greater than 2, and n is an integer greater than 1.

[0217] As an optional implementation, the data acquisition module 1100 is specifically used for:

[0218] Acquire methylation data corresponding to a target methylation site of a target object, and determine the methylation data set according to the methylation data corresponding to the target methylation site;

[0219] The target methylation site is determined based on the degree of difference in methylation data of the same methylation site corresponding to different cancer staging values, or the correlation between the methylation data corresponding to the methylation site and different cancer staging values.

[0220] As an optional implementation, the staging prediction model further includes a standardization model, and the standardization model is used to standardize the methylation data in the methylation data set by means of de-meaning and variance normalization.

[0221] As an optional implementation, the staging prediction model further includes a dimensionality reduction model, and the dimensionality reduction model is used to perform dimensionality reduction processing on the methylation data in the methylation data set according to a PCA method.

[0222] As an optional implementation, the stage prediction module 1101 is specifically used to determine the stage prediction model in the following manner:

[0223] Using the data set to independently train the m classification models, respectively, to determine the classification accuracy of each of the m classification models;

[0224] According to the classification accuracy, k classification models are selected from the m classification models.

[0225] As an optional implementation, after selecting k classification models from m classification models according to the classification accuracy, the stage prediction module 1101 is further used to:

[0226] Using a cross-validation method, the n classification models are selected from the k classification models, wherein m≥k≥n; and / or,

[0227] The model parameters of the screened classification model are adjusted using the grid retrieval method.

[0228] As an optional implementation, the stage prediction module 1101 is specifically used for:

[0229] Using the data set, cross-validating the k classification models respectively to obtain evaluation index values ​​of the k classification models;

[0230] According to the evaluation index values ​​of the k classification models, n classification models are selected from the k classification models.

[0231] As an optional implementation, after acquiring the data set, the data acquisition module 1100 is further configured to:

[0232] The methylation data corresponding to the target methylation site is screened out from the methylation data contained in the data set, and the m classification models are independently trained using the screened methylation data.

[0233] As an optional implementation, the data set includes sample sets of different target objects, the sample sets include methylation data corresponding to each methylation site, and one sample set corresponds to one cancer stage value; the data acquisition module 1100 is specifically used to determine the target methylation site in the following manner:

[0234] According to the cancer staging values ​​corresponding to the sample sets, the data set is divided into a plurality of subsets, wherein each subset corresponds to a cancer staging value, and the sample sets included in different subsets correspond to different cancer staging values;

[0235] For each methylation site, a target methylation site is determined according to the methylation data in multiple subsets respectively corresponding to the methylation site.

[0236] As an optional implementation, the data acquisition module 1100 is specifically used for:

[0237] If the data set is divided into two subsets, then for each methylation site, the target methylation site is determined according to the difference between the methylation data corresponding to the methylation site in the two subsets; or,

[0238] If the data set is divided into two or more subsets, for each methylation site, a target methylation site is determined according to the correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset.

[0239] As an optional implementation, the data acquisition module 1100 is specifically used to determine the difference degree of the methylation data corresponding to the methylation sites in two subsets respectively in the following manner:

[0240] Performing a rank sum test on the methylation data corresponding to the methylation sites in the two subsets respectively, and determining the degree of difference according to the rank sum test value; or,

[0241] The correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset is determined in the following manner:

[0242] A chi-square test is performed on the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset, and the correlation is determined according to the chi-square test value.

[0243] As an optional implementation manner, the multiple subsets include more than two subsets; the data acquisition module 1100 is further specifically used for:

[0244] If, among the multiple divided subsets, there is a subset whose number of sample sets is lower than a threshold, the subset is merged into another subset according to the cancer stage value corresponding to the subset, and the cancer stage value corresponding to the merged another subset is determined, so as to use the merged subsets to determine the target methylation site and train the classification model; wherein the cancer stage value corresponding to the another subset is closest to the cancer stage value corresponding to the subset.

[0245] As an optional implementation manner, the data acquisition module 1100 is further specifically used for:

[0246] For the sample set of the target object, the methylation data corresponding to the methylation sites of the adjacent normal tissues in the sample set and / or the methylation data corresponding to the methylation sites containing missing values ​​are removed.

[0247] As an optional implementation, the stage prediction model is integrated according to the following method:

[0248] Using the outputs of n-1 classification models as the input of the nth classification model, and integrating the n classification models into the stage prediction model;

[0249] The input of the n-1 classification models is the data set, and the output of the nth classification model is the cancer stage value corresponding to the data set.

[0250] Based on the same inventive concept, the embodiment of the present disclosure also provides an electronic device. Since the electronic device is the electronic device in the method in the embodiment of the present disclosure, and the principle of solving the problem by the electronic device is similar to that of the method, the implementation of the electronic device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0251] like Fig.12 As shown, the electronic device includes a processor 1200 and a memory 1201, wherein the memory 1201 is used to store a program executable by the processor 1200, and the processor 1200 is used to read the program in the memory 1201 and perform the following steps:

[0252] Obtain a methylation dataset of a target object;

[0253] Inputting the methylation dataset into a staging prediction model, and outputting a cancer staging value of the target object;

[0254] Among them, the stage prediction model is obtained by integrating n classification models, the model parameters of the n classification models are optimized by using the grid retrieval method, the n classification models are screened from m classification models by using the cross-validation method, and the m classification models are independently trained using the same data set, and m>n, m is an integer greater than 2, and n is an integer greater than 1.

[0255] As an optional implementation manner, the processor 1200 is specifically configured to execute:

[0256] Acquire methylation data corresponding to a target methylation site of a target object, and determine the methylation data set according to the methylation data corresponding to the target methylation site;

[0257] The target methylation site is determined based on the degree of difference in methylation data of the same methylation site corresponding to different cancer staging values, or the correlation between the methylation data corresponding to the methylation site and different cancer staging values.

[0258] As an optional implementation, the staging prediction model further includes a standardization model, and the standardization model is used to standardize the methylation data in the methylation data set by means of de-meaning and variance normalization.

[0259] As an optional implementation, the staging prediction model further includes a dimensionality reduction model, and the dimensionality reduction model is used to perform dimensionality reduction processing on the methylation data in the methylation data set according to a PCA method.

[0260] As an optional implementation, the processor 1200 is specifically configured to determine the staging prediction model in the following manner:

[0261] Using the data set to independently train the m classification models, respectively, to determine the classification accuracy of each of the m classification models;

[0262] According to the classification accuracy, k classification models are selected from the m classification models.

[0263] As an optional implementation manner, after selecting k classification models from m classification models according to the classification accuracy, the processor 1200 is further configured to execute:

[0264] Using a cross-validation method, the n classification models are selected from the k classification models, wherein m≥k≥n; and / or,

[0265] The model parameters of the screened classification model are adjusted using the grid retrieval method.

[0266] As an optional implementation manner, the processor 1200 is specifically configured to execute:

[0267] Using the data set, cross-validating the k classification models respectively to obtain evaluation index values ​​of the k classification models;

[0268] According to the evaluation index values ​​of the k classification models, n classification models are selected from the k classification models.

[0269] As an optional implementation, after acquiring the data set, the processor 1200 is further configured to execute:

[0270] The methylation data corresponding to the target methylation site is screened out from the methylation data contained in the data set, and the m classification models are independently trained using the screened methylation data.

[0271] As an optional implementation, the data set includes sample sets of different target objects, the sample sets include methylation data corresponding to each methylation site, and one sample set corresponds to one cancer staging value; the processor 1200 is specifically configured to determine the target methylation site in the following manner:

[0272] According to the cancer staging values ​​corresponding to the sample sets, the data set is divided into a plurality of subsets, wherein each subset corresponds to a cancer staging value, and the sample sets included in different subsets correspond to different cancer staging values;

[0273] For each methylation site, a target methylation site is determined according to the methylation data in multiple subsets respectively corresponding to the methylation site.

[0274] As an optional implementation manner, the processor 1200 is specifically configured to execute:

[0275] If the data set is divided into two subsets, then for each methylation site, the target methylation site is determined according to the difference between the methylation data corresponding to the methylation site in the two subsets; or,

[0276] If the data set is divided into two or more subsets, for each methylation site, a target methylation site is determined according to the correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset.

[0277] As an optional implementation, the processor 1200 is specifically configured to determine the difference degree of the methylation data corresponding to the methylation sites in two subsets respectively in the following manner:

[0278] Performing a rank sum test on the methylation data corresponding to the methylation sites in the two subsets respectively, and determining the degree of difference according to the rank sum test value; or,

[0279] The correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset is determined in the following manner:

[0280] A chi-square test is performed on the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset, and the correlation is determined according to the chi-square test value.

[0281] As an optional implementation, the multiple subsets include more than two subsets; the data set is divided into multiple subsets according to the cancer stage values ​​corresponding to the sample set, and the processor 1200 is specifically configured to execute:

[0282] If, among the multiple divided subsets, there is a subset whose number of sample sets is lower than a threshold, the subset is merged into another subset according to the cancer stage value corresponding to the subset, and the cancer stage value corresponding to the merged another subset is determined, so as to use the merged subsets to determine the target methylation site and train the classification model; wherein the cancer stage value corresponding to the another subset is closest to the cancer stage value corresponding to the subset.

[0283] As an optional implementation manner, the processor 1200 is further configured to execute:

[0284] For the sample set of the target object, the methylation data corresponding to the methylation sites of the adjacent normal tissues in the sample set and / or the methylation data corresponding to the methylation sites containing missing values ​​are removed.

[0285] As an optional implementation, the stage prediction model is integrated according to the following method:

[0286] Using the outputs of n-1 classification models as the input of the nth classification model, and integrating the n classification models into the stage prediction model;

[0287] The input of the n-1 classification models is the data set, and the output of the nth classification model is the cancer stage value corresponding to the data set.

[0288] Based on the same inventive concept, an embodiment of the present disclosure provides a computer storage medium, the computer storage medium comprising: a computer program code, when the computer program code is executed on a computer, the computer executes any of the cancer staging methods discussed above. Since the principle of solving the problem by the above computer storage medium is similar to that of the cancer staging method, the implementation of the above computer storage medium can refer to the implementation of the method, and the repeated parts will not be repeated.

[0289] In the specific implementation process, the computer storage medium may include: Universal Serial Bus Flash Drive (USB), mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other storage media that can store program codes.

[0290] Based on the same inventive concept, the embodiment of the present disclosure also provides a computer program product, which includes: computer program code, when the computer program code is run on a computer, the computer executes any of the cancer staging methods discussed above. Since the principle of solving the problem by the above computer program product is similar to that of the cancer staging method, the implementation of the above computer program product can refer to the implementation of the method, and the repeated parts will not be repeated.

[0291] The computer program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more conductors, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0292] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0293] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that has the functions specified in one or more boxes.

[0294] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0295] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0296] Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for staging cancer, wherein: The method includes: Obtain a methylation dataset of a target object; Inputting the methylation dataset into a staging prediction model, and outputting a cancer staging value of the target object; Among them, the stage prediction model is obtained by integrating n classification models, the model parameters of the n classification models are optimized by using the grid retrieval method, the n classification models are screened from m classification models by using the cross-validation method, and the m classification models are independently trained using the same data set, and m>n, m is an integer greater than 2, and n is an integer greater than 1.

2. The method according to claim 1, wherein: Obtain the methylation dataset of the target object, including: Acquire methylation data corresponding to a target methylation site of a target object, and determine the methylation data set according to the methylation data corresponding to the target methylation site; The target methylation site is determined based on the degree of difference in methylation data of the same methylation site corresponding to different cancer stage values, or the correlation between the methylation data corresponding to the methylation site and different cancer stage values.

3. The method according to claim 1, wherein: The staging prediction model also includes a standardization model, which is used to standardize the methylation data in the methylation data set by removing the mean and normalizing the variance.

4. The method according to claim 1, wherein: The staging prediction model also includes a dimensionality reduction model, which is used to perform dimensionality reduction processing on the methylation data in the methylation data set according to the PCA method.

5. The method according to claim 1, wherein: The staging prediction model is determined by: Using the data set to independently train the m classification models, respectively, to determine the classification accuracy of each of the m classification models; According to the classification accuracy, k classification models are selected from the m classification models.

6. The method according to claim 5, wherein: After selecting k classification models from m classification models according to the classification accuracy, the method further includes: Using a cross-validation method, the n classification models are selected from the k classification models, wherein m≥k≥n; and / or, The model parameters of the screened classification model are adjusted using the grid retrieval method.

7. The method according to claim 6, wherein: The method of using cross validation to select the n classification models from the k classification models includes: Using the data set, cross-validating the k classification models respectively to obtain evaluation index values ​​of the k classification models; According to the evaluation index values ​​of the k classification models, n classification models are selected from the k classification models.

8. The method according to claim 5, wherein: After obtaining the data set, the method further includes: The methylation data corresponding to the target methylation site is screened out from the methylation data contained in the data set, and the m classification models are independently trained using the screened methylation data.

9. The method according to claim 2 or 8, wherein: The data set includes sample sets of different target objects, the sample sets include methylation data corresponding to each methylation site, and one sample set corresponds to one cancer stage value; the target methylation site is determined by: According to the cancer staging values ​​corresponding to the sample sets, the data set is divided into a plurality of subsets, wherein each subset corresponds to a cancer staging value, and the sample sets included in different subsets correspond to different cancer staging values; For each methylation site, a target methylation site is determined according to the methylation data in multiple subsets respectively corresponding to the methylation site.

10. The method according to claim 9, wherein: Determining the target methylation site according to the methylation data respectively corresponding to the methylation site in the plurality of subsets includes: If the data set is divided into two subsets, then for each methylation site, the target methylation site is determined according to the difference between the methylation data corresponding to the methylation site in the two subsets; or, If the data set is divided into two or more subsets, for each methylation site, a target methylation site is determined according to the correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset.

11. The method according to claim 10, wherein: The difference degree of the methylation data corresponding to the methylation sites in the two subsets is determined by: Performing a rank sum test on the methylation data corresponding to the methylation sites in the two subsets respectively, and determining the degree of difference according to the rank sum test value; or, The correlation between the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset is determined in the following manner: A chi-square test is performed on the methylation data corresponding to the methylation site in each subset and the cancer stage value corresponding to the subset, and the correlation is determined according to the chi-square test value.

12. The method according to claim 9, wherein: The plurality of subsets include more than two subsets; The step of dividing the data set into a plurality of subsets according to the cancer staging values ​​corresponding to the sample set further includes: If, among the multiple divided subsets, there is a subset whose number of sample sets is lower than a threshold, the subset is merged into another subset according to the cancer stage value corresponding to the subset, and the cancer stage value corresponding to the merged another subset is determined, so as to use the merged subsets to determine the target methylation site and train the classification model; wherein the cancer stage value corresponding to the another subset is closest to the cancer stage value corresponding to the subset.

13. The method according to claim 9, wherein: Also includes: For the sample set of the target object, the methylation data corresponding to the methylation sites of the adjacent normal tissues in the sample set and / or the methylation data corresponding to the methylation sites containing missing values ​​are removed.

14. The method according to claim 1, wherein: The stage prediction model is integrated according to the following method: Using the outputs of n-1 classification models as the input of the nth classification model, and integrating the n classification models into the stage prediction model; The input of the n-1 classification models is the data set, and the output of the nth classification model is the cancer stage value corresponding to the data set.

15. An electronic device, wherein: The electronic device comprises a processor and a memory, wherein the memory is used to store a program executable by the processor, and the processor is used to read the program in the memory and execute the steps of any one of the methods of claims 1 to 14.

16. A computer storage medium having a computer program stored thereon, wherein: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 14 are implemented.

Images