Multi-omics constrained methylated CPG site screening method, equipment and program product

Abstract

The invention relates to the field of intelligent medical treatment, in particular to a methylated CPG site screening method and device for constraining multiple omics and a program product. Comprising the following steps: acquiring transcription data, CNV data and a methylated CPG probe of a sample to be detected; processing the methylated CPG probe, the transcription factor and the CNV data, and performing regression calculation to obtain a regression methylated CPG probe, a transcription factor, a conversion transcription factor, CNV data and a differential expression gene; and performing gene variation analysis on the basis of the regression methylated CPG probe, the transcription factor, the conversion transcription factor, the CNV data and the differential expression gene to obtain a methylated CPG site. The methylation site screening provided by the invention considers the characteristics of multiple omics, and has good clinical value.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical care, and specifically to a method, device, program product and computer-readable storage medium for screening methylated CPG sites using constrained multi-omics. Background Art

[0002] Screening of methylated CpG sites is an important part of epigenetic research, involving a variety of techniques and methods. Each method has its own unique advantages and limitations. For example, whole genome methylation sequencing (WGBS) can detect the methylation status of every CpG site in the entire genome, providing the most comprehensive data. However, WGBS requires a high sequencing depth, which makes the experimental cost very expensive and not suitable for large-scale sample research. Then the bisulfite array was proposed, which can detect thousands of specific CpG sites at the same time, suitable for large-scale sample research, but it depends on the design of commercial arrays, may miss some important methylation sites, and the cost is high. Methylation-specific PCR (MSP) and reduced representation bisulfite sequencing (RRBS) are cheaper, but PCR requires the design of specific primers, and the requirements for primer design are high. Only a few CpG sites can be detected at a time. RRBS has limited coverage and mainly detects high CpG density areas. The detection ability of low CpG density areas is weak. Each method has its applicable scenarios and limitations, but none of them has ever considered methylation site screening from multiple dimensions. Summary of the invention

[0003] Methylated CpG site screening combined with multi-omics is a method that integrates multiple data types, aiming to fully understand the mechanisms of gene expression regulation and disease occurrence and development from multiple perspectives. Multi-omics data usually include genomics, transcriptomics, proteomics, metabolomics, etc. By integrating these data, the function and biological significance of methylated CpG sites can be more accurately identified and verified. Therefore, in response to the above problems, the present invention provides a constrained multi-omics methylation CPG site screening method, which specifically includes: S1, obtaining gene expression data of the sample to be tested, including transcription data, CNV data, and methylation CPG probes; S2, performing difference calculation on the transcription data to obtain differentially expressed genes; S3, performing correlation calculation on the transcription data to obtain the correlation of transcription factors; S4, when the correlation of the transcription factor is negatively correlated, assigning a negative value to the negatively correlated transcription factor to obtain a conversion transcription factor; S5, assigning a negative value to the methylation CPG probe to obtain a conversion methylation CPG probe; S6, performing regression calculation on the conversion methylation CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes to obtain regressed methylation CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes; S7, performing gene variation analysis based on the regressed methylation CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes to obtain methylation CPG sites.

[0004] The regression calculation is performed by a regression model, including one or more of the following: Lasso regression, ridge regression, Bayesian linear regression, elastic network regression; Optionally, the regression calculation obtains an optimal regularization parameter through ridge regression, and the regression coefficients are ranked based on the optimal regularization parameter to obtain the first-ranked regression coefficient, as well as the first-ranked corresponding methylated CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes.

[0005] Optionally, the optimal regularization parameter is a value interval of a regularization parameter generated first, and the regularization parameter is optimized in the value interval to obtain the optimal regularization parameter.

[0006] Optionally, the parameter optimization is performed by one or more of the following: grid search method, random search method, particle swarm optimization algorithm, niche algorithm, simulated annealing algorithm, genetic algorithm, ant colony optimization algorithm.

[0007] The transcription factors are subjected to correlation calculation to obtain a correlation ranking, and the top N transcription factors in the ranking are retained, where N is a natural number greater than 2.

[0008] Optionally, the correlation calculation uses Pearson correlation coefficient to calculate and obtain correlation ranking.

[0009] Optionally, the methylated CPG probes are screened to obtain screened methylated CPG probes, and the screened methylated CPG probes are converted to negative values ​​to obtain converted methylated CPG probes; wherein the screening is to screen probes whose methylation is higher than a preset threshold or whose methylation is lower than a preset threshold.

[0010] Optionally, the methylation screening also includes similarity screening, performing similarity calculation on the screened methylated CPG probes to obtain similar methylated CPG probes, and performing negative value conversion on the similar methylated CPG probes to obtain converted methylated CPG probes; optionally, the similar methylated CPG probes are the first L similar methylated CPG probes sorted by similarity, where L is a natural number greater than 2.

[0011] Optionally, the negation is multiplication by negative one.

[0012] Optionally, the method further includes assigning negative values ​​to the differentially expressed genes, calculating the correlation of the differentially expressed genes, assigning negative values ​​to the negatively correlated differentially expressed genes to obtain converted differentially expressed genes; and performing regression calculation on the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data, differentially expressed genes, and converted differentially expressed genes to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, differentially expressed genes, and converted differentially expressed genes.

[0013] Optionally, the method also includes multi-scale conversion, normalizing the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data, and differentially expressed genes to obtain normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, and normalized differentially expressed genes, and performing regression calculation on the normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, and normalized differentially expressed genes to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, and differentially expressed genes.

[0014] The method further includes data preprocessing, normalizing the gene expression data to obtain processed gene expression data, and performing processing of S2-S7 based on the processed gene expression data.

[0015] Optionally, the CNV data includes data in SEG MEAN format.

[0016] Optionally, the methylated CPG probe comprises a beta value.

[0017] The object of the present invention is to provide a glioma prediction method based on H3 K27 mutation, comprising: obtaining gene data of a sample to be tested; extracting methylation site features of the gene data; the methylation site features are obtained according to the above-mentioned constrained multi-omics methylation CPG site screening method; and predicting based on the methylation site features to obtain a prediction result that the sample to be tested is pontine-like or thalamic-like.

[0018] The pontine-like tumor is located in the pons, and the thalamus-like tumor is located outside the pons. The characteristics also include mutation characteristics, which extract mutation characteristics of gene data, and predict based on the methylation site characteristics and mutation characteristics to obtain a prediction result of whether the sample to be tested is pontine-like or thalamus-like.

[0019] Optionally, the mutation characteristics include one or more of the following: TP53 mutation, NF1 mutation.

[0020] Optionally, the characteristics also include clinical characteristics, clinical data are acquired, clinical characteristics in the clinical data are extracted, and a prediction result is obtained based on the methylation site characteristics and clinical characteristics that the sample to be tested is pontine-like or thalamic-like.

[0021] Optionally, the clinical characteristics include one or more of the following: age, MKI67 IHC index.

[0022] Optionally, the characteristics include clinical characteristics and mutation characteristics, and a prediction result is obtained based on the methylation site characteristics, mutation characteristics, and clinical characteristics to determine whether the sample to be tested is pontine-like or thalamic-like.

[0023] Optionally, the method also includes data processing, performing data dimensionality reduction on the methylation features to obtain reduced dimensionality data, and then sorting the reduced dimensionality data to obtain the first S reduced dimensionality methylation features, where S is a natural number greater than 3, and predicting based on the reduced dimensionality methylation site features to obtain a prediction result that the sample to be tested is pontine-like or thalamic-like.

[0024] Optionally, the mutation signature or clinical signature is screened based on edge feature selection.

[0025] The prediction is made through one or more of the following models: Naive Bayes, Support Vector Machine, Random Forest, Decision Tree, Convolutional Neural Network, Transformer.

[0026] Optionally, the prediction is performed by inputting the features into respective naive Bayes to obtain respective prediction probabilities, and voting on the respective prediction probabilities to obtain a final prediction result.

[0027] The object of the present invention is to provide a computer program product, which includes a computer program or instructions, and the computer program or instructions are executed by a processor to implement the above-mentioned constrained multi-omics methylation CPG site screening method, or to implement the above-mentioned glioma prediction method based on H3 K27 mutation.

[0028] The object of the present invention is to provide a computer device, comprising a memory, a processor and a computer program or instructions stored on the memory, wherein the computer program or instructions are executed by the processor to implement the above-mentioned constrained multi-omics methylation CPG site screening method, or to implement the above-mentioned glioma prediction method based on H3 K27 mutation.

[0029] The object of the present invention is to provide a computer-readable storage medium having a computer program or instruction stored thereon, wherein the computer program or instruction is executed by a processor to implement the above-mentioned constrained multi-omics methylation CPG site screening method, or to implement the above-mentioned glioma prediction method based on H3 K27 mutation.

[0030] Advantages of the present invention: 1. The present invention proposes a constrained multi-omics methylation CPG site screening method, which combines transcription data, gene data, copy variant number data, and methylation data. The methylation CPG sites screened by constrained multi-omics features are more in line with the working environment of the gene regulatory network than the single methylation data screening method, thereby improving the accuracy of site screening.

[0031] 2. In the methylation CPG screening sites of constrained multi-omics, the multi-omics data are integrated through the regression model to obtain the multi-omics data corresponding to the optimal regularization parameter, and then the methylation CPG sites are obtained based on the analysis of the multi-omics data obtained by this screening. The optimal regularization parameter is obtained by optimizing the optimization parameter.

[0032] 3. Diffuse midline glioma (DMG) with H3 K27 changes is a highly lethal tumor of the central nervous system, but there is a lack of effective treatment methods for glioma. Studies have found that H3 K27 changes alone are not sufficient to cause tumorigenesis, and tumor transformation requires at least one partner change. Animal models have shown that H3 K27-changed tumors with different partner changes exhibit different growth rates, diffusion and invasion phenotypes. For this reason, based on data analysis of gliomas with H3 K27 changes, the present invention proposes a new classification, pontine-like and thalamic-like, and predicts different subtypes of gliomas by using methylated CPG sites obtained through constrained multi-omics data screening, which is helpful for early prediction, warning, intervention, diagnosis and treatment of the disease.

[0033] 4. The classification prediction of glioma also includes mutation characteristics and clinical features, and predictions are made from multiple dimensions to further improve the accuracy and reliability of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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 invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 A schematic diagram of a process flow of a method for screening methylated CPG sites using constrained multi-omics provided in an embodiment of the present invention; Figure 2 A schematic diagram of a methylated CPG site screening system for constrained multi-omics provided in an embodiment of the present invention; Figure 3 A schematic diagram of a methylated CPG site screening device for constrained multi-omics provided in an embodiment of the present invention; Figure 4 A flow chart of CM3R provided by an embodiment of the present invention; Figure 5 Data analysis results of glioma provided in the embodiments of the present invention, (B) Clinical information, data availability and somatic mutations of H3-DSGs included in this study. C / T / L represent the cervical, thoracic and lumbar segments of the human spinal cord, respectively. Summarize the diffuse midline glioma driver genes identified by MutSigCV and the reported key mutations. (C) cSNF network showing two multi-omics subtypes of H3-DSGs. TP53 and NF1 mutation status are indicated by edge color. Edge width indicates the strength of evidence supporting patient-patient similarity from multi-omics data. Lymph node size is positively correlated with the age of the patient at diagnosis (7 to 61 years old). Node color indicates the H3-DSG subtype defined by the cSNF method. (D) Mutation frequency of key driver genes in different diseases. Brain-M-Pons and brain-M-Medulla are from Chen et al. (PMID: 32555164). (E) Diagnosis age of patients in different disease groups. (F) Kaplan-Meier survival curves of tumor patients in different groups;

[0036] Figure 6 Functional annotation of differentially expressed genes between pontine-like and thalamic-like H3-DSGs provided in the embodiments of the present invention (A) Volcano plot showing differentially expressed genes between the two tumor subtypes in bulk RNA-seq; (B) Comparison of tumor purity between the two subtypes calculated based on WES data; (C and D) Comparison of lymphocyte infiltration and myeloid infiltration between pontine-like and thalamic-like subgroups by RNA-seq data; Figure 7(A) The proportion of cycling cells in pontine-like and thalamic-like H3-DSGs; (B) Immunohistochemistry of Ki-67 (left) and Ki-67 index in pontine-like and thalamic-like H3-DSGs provided in the embodiments of the present invention; Figure 8 Comparison of CNVs / LOH events calculated from two subgroups of tumor WES data provided in an embodiment of the present invention; Fig. 9 The expression of FANCI protein in two groups of tumors provided in the examples of the present invention, scale bar 20 μm. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0038] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as S101, S102, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0039] Figure 1 A schematic diagram of a method for screening methylated CPG sites by constrained multi-omics provided in an embodiment of the present invention specifically includes: S1: Obtain gene expression data of the sample to be tested, including transcription data, CNV data, and methylation CPG probes; In one embodiment, the method further includes data preprocessing, normalizing the gene expression data to obtain processed gene expression data, and performing processing of S2-S7 based on the processed gene expression data.

[0040] In one embodiment, the CNV data includes data in a SEG MEAN format.

[0041] In one embodiment, the methylated CPG probe comprises a beta value.

[0042] In a specific embodiment, a novel method of CM3R (constrained multi-omics integrated regression-based regulatory factor ranking algorithm) can identify copy number variations, transcription factors, and DNA methylation CpG sites that are highly correlated with subtype differential gene expression. The method includes 3 steps, (a) data preparation, (b) TFs extraction and CpG probe pre-selection, and (c) multi-scale normalization and regression, see Figure 4 As shown in the figure. In the data preparation of the CM3R method, gene expression data undergo a series of normalization procedures before being used as input, including quantile normalization and minimum and maximum normalization. At the same time, CNV data in the 'segmean' format and beta values ​​of CpG probes are required.

[0043] S2: differentially expressed genes were obtained by differential calculation of transcription data; In one embodiment, the method further includes assigning negative values ​​to the differentially expressed genes, calculating the correlation of the differentially expressed genes, assigning negative values ​​to the negatively correlated differentially expressed genes to obtain converted differentially expressed genes; the positively correlated differentially expressed genes remain unchanged, and the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data, (positively correlated) differentially expressed genes, and converted differentially expressed genes are subjected to regression calculation to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, (positively correlated) differentially expressed genes, and converted differentially expressed genes.

[0044] S3: Correlation calculation of transcription data was performed to obtain the correlation of transcription factors; In one embodiment, the transcription factors are subjected to correlation calculation to obtain a correlation ranking, and the top N transcription factors in the ranking are retained, where N is a natural number greater than 2.

[0045] In one embodiment, the correlation calculation uses the Pearson correlation coefficient to calculate and obtain the correlation ranking.

[0046] In a specific embodiment, in (b) transcription factor extraction and cpg probe pre-selection, the transcription factors (TFs) of each gene are first extracted from the RegNetwork database and further selected based on the Pearson correlation coefficient index. In the CM3R method, up to 3 transcription factors are considered for each gene.

[0047] S4: When the correlation of transcription factors is negative, negative values ​​are assigned to the negatively correlated transcription factors to obtain conversion transcription factors; In a specific embodiment, two additional transformations are performed on the beta values ​​of these selected CpG probes and the gene expression of transcription factors. The beta values ​​of all selected CpG probes are multiplied by (-1). For transcription factor gene expression that shows a negative correlation with the current differentially expressed (DE) genes, multiplication by (-1) is also applied to reverse their values.

[0048] S5: assigning a negative value to the methylated CPG probe to convert the methylated CPG probe; In one embodiment, the methylated CPG probes are screened to obtain screened methylated CPG probes, and the screened methylated CPG probes are converted to negative values ​​to obtain converted methylated CPG probes; wherein the screening is to screen probes whose methylation is higher than a preset threshold or whose methylation is lower than a preset threshold.

[0049] In one embodiment, the methylation screening further includes similarity screening, performing similarity calculation on the screened methylated CPG probes to obtain similar methylated CPG probes, and performing negative value conversion on the similar methylated CPG probes to obtain converted methylated CPG probes.

[0050] Optionally, the similar methylated CPG probes are the first L similar methylated CPG probes sorted by similarity, where L is a natural number greater than 2.

[0051] Optionally, the negation is multiplication by negative one.

[0052] In a specific embodiment, the methylated CpG probes are selected by first selecting CpG probes that show high or low methylation patterns in the entire gene body and the upstream 2000b of the transcription start site (TSS) region, and further adopting a selection procedure similar to TF.

[0053] In a specific embodiment, a multi-scale normalization technique is used to appropriately scale gene expression data (including positively / negatively correlated transcription factors and differentially expressed genes), CNV data, and CpG beta values ​​using a sigmoid transformation.

[0054] S6: Regression calculation is performed on the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data and differentially expressed genes to obtain the regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data and differentially expressed genes.

[0055] In one embodiment, the regression calculation is calculated by a regression model, including one or more of the following: Lasso regression, ridge regression, Bayesian linear regression, elastic network regression. In one embodiment, the regression calculation obtains an optimal regularization parameter by ridge regression, and the regression coefficients are sorted based on the optimal regularization parameter to obtain the first-ranked regression coefficient, as well as the first-ranked corresponding methylated CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes.

[0056] In one embodiment, the optimal regularization parameter is to generate a value interval of the regularization parameter first, and to perform regularization parameter optimization in the value interval to obtain the optimal regularization parameter; Optionally, the parameter optimization is performed by one or more of the following: grid search method, random search method, particle swarm optimization algorithm, niche algorithm, simulated annealing algorithm, genetic algorithm, ant colony optimization algorithm.

[0057] In one embodiment, the method also includes multi-scale conversion, normalizing the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data, and differentially expressed genes to obtain normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, and normalized differentially expressed genes, and performing regression calculation on the normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, and normalized differentially expressed genes to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, and differentially expressed genes.

[0058] In one embodiment, the method also includes multi-scale conversion, normalizing the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data, differentially expressed genes, and converted differentially expressed genes to obtain normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, normalized differentially expressed genes, and normalized converted differentially expressed genes, regressing the normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, normalized converted differentially expressed genes, and normalized differentially expressed genes to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, converted differentially expressed genes, and differentially expressed genes, and performing gene difference analysis based on the regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, converted differentially expressed genes, and differentially expressed genes to obtain methylated CPG sites.

[0059] In a specific embodiment, a restricted ridge regression model is used to simulate the relationship between the gene expression of the current differentially expressed (DE) genes and independent variables, including CNV data, selected CpG probes and transcription factors, which forces the coefficients of the regulators to be non-negative.

[0060] During the regression phase of the CM3R method, a range of values ​​for the regularization parameter (denoted as α) is systematically employed. These values ​​are varied to optimize the performance of the model and achieve the best fit to the given data. By exploring different values ​​of α, the CM3R method can find the optimal regularization parameter that balances model complexity and prediction accuracy. Once the optimal α value is determined, the CM3R method assigns the top-ranked regulatory factor type to the current DE gene based on the ranking of the regression coefficients, that is, the top-ranked transcription factors (TFs), top-ranked CpGs, and top-ranked CNVs associated with the DE gene are determined.

[0061] S7: Based on the regressed methylated CPG probes, transcription factors, conversion transcription factors, CNV data and differentially expressed genes, gene variation analysis is performed to obtain methylated CPG sites.

[0062] In one embodiment, the analysis results are obtained by performing gene variable analysis through GSVA.

[0063] The disclosed embodiments of the present invention further provide a computer program product or system, including a computer program, which, when executed by a processor, implements the above-mentioned constrained multi-omics methylation CPG site screening method steps.

[0064] Figure 2 A schematic diagram of a constrained multi-omics methylation CPG site screening system provided in an embodiment of the present invention specifically includes: an acquisition unit: acquiring gene expression data of a sample to be tested, including transcription data, CNV data, and methylation CPG probes; Difference unit: differentially calculated transcription data to obtain differentially expressed genes; Transcription unit: Correlation calculation of transcription data is performed to obtain the correlation of transcription factors; Conversion unit: when the correlation of transcription factors is negative, negative values ​​are assigned to the negatively correlated transcription factors to obtain conversion transcription factors; Methylation unit: assign a negative value to the methylated CPG probe to convert the methylated CPG probe; Regression unit: performing regression calculation on the converted methylation CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes to obtain the regressed methylation CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes; Analysis unit: performing gene variation analysis based on the regressed methylated CPG probes, transcription factors, conversion transcription factors, CNV data and differentially expressed genes to obtain methylated CPG sites.

[0065] Figure 3A schematic diagram of a methylated CPG site screening device for constrained multi-omics provided by an embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned constrained multi-omics methylation CPG site screening methods is performed.

[0066] The disclosed embodiments of the present invention further provide a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor by any one of the above-mentioned constrained multi-omics methylation CPG site screening methods.

[0067] An embodiment of the present invention provides a glioma prediction method based on H3 K27 mutation, comprising: obtaining gene data of a sample to be tested; extracting methylation site features of the gene data; the methylation site features are obtained by the above-mentioned constrained multi-omics methylation CPG site screening method; and based on the methylation site features, a prediction is made to obtain a prediction result that the sample to be tested is pontine-like or thalamic-like.

[0068] In one embodiment, the pontine-like tumor is located in the pons, and the thalamic-like tumor is located outside the pons.

[0069] In one embodiment, the feature also includes a mutation feature, and the mutation feature of the gene data is extracted. Based on the methylation site feature and the mutation feature, a prediction is made to obtain a prediction result that the sample to be tested is pontine-like or thalamic-like.

[0070] In one embodiment, the mutation characteristics include one or more of the following: TP53 mutation, NF1 mutation.

[0071] In one embodiment, the features also include clinical features, clinical data are acquired, clinical features in the clinical data are extracted, and a prediction result is obtained based on the methylation site features and clinical features to determine whether the sample to be tested is pontine-like or thalamic-like.

[0072] In one embodiment, the clinical characteristics include one or more of the following: age, MKI67 IHC index.

[0073] In one embodiment, the characteristics include clinical characteristics and mutation characteristics, and a prediction result is obtained based on the methylation site characteristics, mutation characteristics, and clinical characteristics to determine whether the sample to be tested is pontine-like or thalamic-like.

[0074] In one embodiment, the method further includes data processing, performing data dimensionality reduction on the methylation features to obtain reduced dimensionality data, and then sorting the reduced dimensionality data to obtain the first S reduced dimensionality methylation features, where S is a natural number greater than 3, and predicting based on the reduced dimensionality methylation site features to obtain a prediction result that the sample to be tested is pontine-like or thalamic-like.

[0075] In one embodiment, the mutation signature or clinical signature is screened based on edge-based signature selection.

[0076] In one embodiment, the prediction is performed by one or more of the following models: Naive Bayes, Support Vector Machine, Random Forest, Decision Tree, Convolutional Neural Network, Transformer; Optionally, the prediction is performed by inputting the features into respective naive Bayes to obtain respective prediction probabilities, and voting on the respective prediction probabilities to obtain a final prediction result.

[0077] In one embodiment, five naive Bayes are constructed for methylation features, mutation features (TP53 mutation, NF1 mutation), and clinical features (age, MKI67 IHC index) and trained in parallel to obtain prediction models for the corresponding features.

[0078] In a specific embodiment, the present invention analyzes multi-omics and spatially resolved single-cell data from 36 H3-DSGs and finds two different subgroups: pontine-like and thalamic-like. Thalamic-like tumors are characterized by NF1 mutations and show more immune infiltration. They contain a large number of proinflammatory myeloid cells and CD8+ cytotoxic T cells, and the patients are older and have a relatively good prognosis. In contrast, pontine-like tumors show abundant TP53 mutations, high proliferation niches, genomic instability, and the patients are older and have a relatively poor prognosis.

[0079] To clarify the subtypes of H3 K27M-mutant spinal cord gliomas, multi-omics clustering and subtype identification were performed on 36 cases with DNA methylation, WES, and RNA sequencing data ( Figure 5 B). There were 22 males and 14 females, ranging in age from 7 to 61 years old, with a median age of 33 years old. The tumors were extensively resected from the first cervical segment (C1) to the first lumbar segment (L1). Two of the cases had multifocal tumors.

[0080] In another specific embodiment, in order to better describe the tumor characteristics of H3-DSG subtype 1 and subtype 2, the present invention compared the somatic mutation spectrum, age at diagnosis and overall survival outcomes of the H3-DSG subgroup with previously reported brain DMG subgroups in an independent cohort consisting of m-pontine and m-medullary brain DMG17. Tumors of the m-pontine DMG subgroup are mainly located in the pons, while tumors of the m-medullary DMG subgroup are mostly located outside the pons. Strikingly, the newly discovered H3-DSG subtype 1 exhibited somatic characteristics similar to those of the m-pontine DMG subgroup, while the somatic characteristics of H3-DSG subtype 2 were similar to those of the m-medullary brain DMG subgroup, with the only exceptions being SF3B1 and FGFR1 mutations, which were more frequent in m-medullary brain DMGs than in H3-DSG subtype 2 ( Figure 5 The age distribution at diagnosis was similar in the H3-DSG subtype 1 and M-Pons brain DMG groups and was significantly lower than that in the H3-DSG subtype 2 (P=4.5e-05) and M-Pons brain DMG groups (P=6.4e-06), respectively ( Figure 5 E). The survival rates of the H3-DSG 1 group and the m-pontine DMG group were significantly lower than those of the H3-DSG 2 group (P = 8.7e-04) and the m-medullary DMG group (P = 1.4e-04), respectively ( Figure 5 F). These results indicate that H3-DSG subtype 1 and H3-DSG subtype 2 have similar molecular and clinical characteristics to m-pontine and m-medullary DMG, respectively. Based on these findings, the present invention designates H3-DSG subtype 1 as pontine-like and H3-DSG subtype 2 as thalamic-like.

[0081] In one embodiment, by studying the differentially expressed genes between pontine-like and thalamic-like H3-DSGs ( Figure 6 A), we found that the genes significantly upregulated in pontine-like tumors were mainly related to cell division and cell cycle-related functions, such as MEX3A, KIF2C, FANCI, etc. ( Figure 6 In contrast, genes upregulated in thalamic-like tumors included immune and inflammatory activities. Thus, thalamic-like tumors had lower tumor purity (P = 0.006) and higher lymphocyte (P = 0.0044) and myeloid scores (P = 0.0006) compared with pontine-like H3-DSGs ( Figure 6BD). In fact, scRNA-seq data showed that pontine-like and thalamic-like H3-DSGs displayed different cell type compositions and tumor cell states. Malignant cells accounted for 80% to 90% of pontine-like tumors, and the remaining components were mostly myeloid cells, with rare T cells. In contrast, in thalamic-like tumors, malignant cells accounted for only about 50%, and the remaining 50% included myeloid cells, T cells, NK cells, neutrophils, endothelial cells, and oligodendrocytes. Consistent with this, immunohistochemistry (IHC) of 28 H3-DSGs showed that there were more CD14-positive myeloid cells in thalamic-like H3-DSGs. Multicolor staining also showed that there were more myeloid and lymphoid cells in thalamic-like H3-DSGs. Correspondingly, CD14 (myeloid), CD3e (T cells), MPO (neutrophils), CD44, HLA-A and HLA-DR (antigen presentation) were more abundant in thalamic-like H3-DSGs than in pontine-like H3-DSGs.

[0082] In a specific embodiment, the single cell analysis results also showed that the proportion of CD8-positive cytotoxic T cells in T cells was significantly increased in thalamus-like H3-DSG. The results of immunohistochemistry showed that the proportion of CD3e-positive cells and GZMB-positive cells in thalamus-like H3-DSG was also significantly higher than that of the corresponding cells in pons-like H3-DSG. This indicates that the proportion of cytotoxic T cells in thalamus-like H3-DSG has increased.

[0083] In a specific embodiment, single cell analysis revealed that more cells in the pontine-like H3-DSG were in a cycle of cell proliferation ( Figure 7 A), and the index of Ki-67, a marker of cell proliferation, was significantly increased in H3-DSG of pontine-like cells ( Figure 7 B). We then found that the copy number variation and heterozygosity loss rates of pontine-like H3-DSG were significantly higher than those of thalamic-like H3-DSG ( Figure 8 ), suggesting that their genome is more unstable. The expression level of FANCI, a protein related to DNA damage repair, was significantly increased in pontine-like H3-DSG ( Fig. 9 ).

[0084] In one specific embodiment, the prediction model: The naïve Bayes algorithm in the naivebayes (v0.9.7) R package was used to build the machine learning model. The original feature pool consisted of several parts: the top 5 individual cell driver mutations identified by the MutsigCV pipeline, the other top 659 cpgs through the CMR3 pipeline, and clinical features such as age and KI67 IHC index.

[0085] For methylation features, principal component analysis (PCA) was first used for dimensionality reduction and data integration, and the first four principal components (PCs) were retained. For other features, edge-based feature selection was used, each feature was input into the Naïve Bayes model, and leave-one-out cross-validation was performed, and finally age, MKI67 IHC, TP53, and NF1 were selected.

[0086] Then, a multi-omics feature soft voting technique based on the Naïve Bayes algorithm was used to assign groups to different samples, trained using leave-one-out cross-validation. Different voting models were used to assign subgroup labels based on available data sources.

[0087] The disclosed embodiments of the present invention further provide a computer program product or system, including a computer program, which, when executed by a processor, implements the above-mentioned steps of the method for predicting glioma based on H3 K27 mutation.

[0088] The glioma prediction device based on H3 K27 mutation provided in the embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned glioma prediction methods based on H3 K27 mutations is executed.

[0089] The disclosed embodiments of the present invention further provide a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to perform any one of the above-mentioned methods for predicting glioma based on H3 K27 mutation.

[0090] The verification results of this verification embodiment show that assigning inherent weights to indications can improve the performance of the method relative to the default setting. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units. A person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above-mentioned embodiments may be completed by instructing the relevant hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0091] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be implemented by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned medium storage can be a read-only memory, a disk or an optical disk, etc.

[0092] The above is a detailed introduction to a computer device provided by the present invention. For a person skilled in the art, according to the concept of the embodiments of the present invention, there may be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for screening methylated CPG sites by constrained multi-omics, characterized in that: include: S1. Obtain gene expression data of the sample to be tested, including transcription data, CNV data, and methylation CPG probes; S2, performing differential calculation on transcription data to obtain differentially expressed genes; S3, performing correlation calculation on the transcription data to obtain the correlation of transcription factors; S4, when the correlation of transcription factors is negative, assign negative values ​​to the negatively correlated transcription factors to obtain conversion transcription factors; S5, assigning a negative value to the methylated CPG probe to convert the methylated CPG probe; S6, performing regression calculation on the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data and differentially expressed genes to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data and differentially expressed genes; S7. Based on the regressed methylated CPG probes, transcription factors, conversion transcription factors, CNV data and differentially expressed genes, gene variation analysis is performed to obtain methylated CPG sites.

2. The method for screening methylated CPG sites by constrained multi-omics according to claim 1, characterized in that: The regression calculation is performed by a regression model, including one or more of the following: Lasso regression, ridge regression, Bayesian linear regression, elastic network regression; Optionally, the regression calculation obtains an optimal regularization parameter by ridge regression, and the regression coefficients are sorted based on the optimal regularization parameter to obtain the first-ranked regression coefficient, as well as the first-ranked corresponding methylated CPG probe, transcription factor, conversion transcription factor, CNV data and differentially expressed genes; Optionally, the optimal regularization parameter is a value interval of a regularization parameter generated first, and the regularization parameter is optimized in the value interval to obtain the optimal regularization parameter; Optionally, the parameter optimization is performed by one or more of the following: grid search method, random search method, particle swarm optimization algorithm, niche algorithm, simulated annealing algorithm, genetic algorithm, ant colony optimization algorithm.

3. The method for screening methylated CPG sites by constrained multi-omics according to claim 1, characterized in that: The transcription factors are subjected to correlation calculation to obtain a correlation ranking, and the top N transcription factors in the ranking are retained, where N is a natural number greater than 2; Optionally, the correlation calculation uses Pearson correlation coefficient to calculate and obtain correlation ranking; Optionally, the methylated CPG probes are screened to obtain screened methylated CPG probes, and the screened methylated CPG probes are converted to negative values ​​to obtain converted methylated CPG probes; The screening is to screen probes with methylation higher than a preset threshold or lower than a preset threshold; Optionally, the methylation screening further comprises similarity screening, performing similarity calculation on the screened methylated CPG probes to obtain similar methylated CPG probes, and performing negative value conversion on the similar methylated CPG probes to obtain converted methylated CPG probes; Optionally, the similar methylated CPG probes are the first L similar methylated CPG probes sorted according to similarity, where L is a natural number greater than 2; Optionally, the negative value is multiplied by negative one; Optionally, the method further comprises assigning negative values ​​to the differentially expressed genes, calculating the correlation of the differentially expressed genes, assigning negative values ​​to the negatively correlated differentially expressed genes to obtain converted differentially expressed genes; performing regression calculation on the converted methylated CPG probe, transcription factor, converted transcription factor, CNV data, differentially expressed gene, and converted differentially expressed gene to obtain regressed methylated CPG probe, transcription factor, converted transcription factor, CNV data, differentially expressed gene, and converted differentially expressed gene; Optionally, the method also includes multi-scale conversion, normalizing the converted methylated CPG probes, transcription factors, converted transcription factors, CNV data, and differentially expressed genes to obtain normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, and normalized differentially expressed genes, and performing regression calculation on the normalized methylated CPG probes, normalized transcription factors, normalized converted transcription factors, normalized CNV data, and normalized differentially expressed genes to obtain regressed methylated CPG probes, transcription factors, converted transcription factors, CNV data, and differentially expressed genes.

4. The method for screening methylated CPG sites by constrained multi-omics according to claim 1, characterized in that: The method further comprises data preprocessing, normalizing the gene expression data to obtain processed gene expression data, and performing processing of S2-S7 based on the processed gene expression data; Optionally, the CNV data includes data in SEG MEAN format; Optionally, the methylated CPG probe comprises a beta value.

5. A method for predicting glioma based on H3 K27 mutation, characterized in that: include: Obtaining genetic data of the sample to be tested; Extracting methylation site features of the gene data; The methylation site characteristics are obtained according to the constrained multi-omics methylation CPG site screening method according to claims 1-4; Based on the methylation site characteristics, a prediction result is obtained that the sample to be tested is pontine-like or thalamic-like.

6. The method for predicting glioma based on H3 K27 mutation according to claim 5, characterized in that: The pontine-like tumor is located in the pons, and the thalamic-like tumor is located outside the pons. Optionally, the feature also includes a mutation feature, the mutation feature of the gene data is extracted, and a prediction result that the sample to be tested is pontine-like or thalamic-like is obtained based on the methylation site feature and the mutation feature; Optionally, the mutation characteristics include one or more of the following: TP53 mutation, NF1 mutation; Optionally, the feature also includes clinical features, clinical data is acquired, clinical features in the clinical data are extracted, and a prediction result that the sample to be tested is pontine-like or thalamic-like is obtained based on the methylation site features and clinical features; Optionally, the clinical characteristics include one or more of the following: age, MKI67 IHC index; Optionally, the characteristics include clinical characteristics and mutation characteristics, and a prediction result of whether the sample to be tested is pontine-like or thalamic-like is obtained based on the methylation site characteristics, mutation characteristics, and clinical characteristics; Optionally, the method further includes data processing, performing data dimensionality reduction on the methylation features to obtain reduced dimensionality data, and then sorting the reduced dimensionality data to obtain the first S reduced dimensionality methylation features, where S is a natural number greater than 3, and predicting based on the reduced dimensionality methylation site features to obtain a prediction result that the sample to be tested is pontine-like or thalamic-like; Optionally, the mutation signature or clinical signature is screened based on edge feature selection.

7. The method for predicting glioma based on H3 K27 mutation according to claim 5, characterized in that: The prediction is made by one or more of the following models: Naive Bayes, Support Vector Machine, Random Forest, Decision Tree, Convolutional Neural Network, Transformer; Optionally, the prediction is performed by inputting the features into respective naive Bayes to obtain respective prediction probabilities, and voting on the respective prediction probabilities to obtain a final prediction result.

8. A computer program product comprising a computer program or instructions, characterized in that: The computer program or instructions are executed by a processor to implement the constrained multi-omics methylation CPG site screening method described in any one of claims 1-4, or to implement the glioma prediction method based on H3 K27 mutation described in any one of claims 5-7.

9. A computer device comprising a memory, a processor and a computer program or instruction stored in the memory, characterized in that: The computer program or instructions are executed by a processor to implement the constrained multi-omics methylation CPG site screening method described in any one of claims 1-4, or to implement the glioma prediction method based on H3 K27 mutation described in any one of claims 5-7.

10. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: The computer program or instructions are executed by a processor to implement the constrained multi-omics methylation CPG site screening method described in any one of claims 1-4, or to implement the glioma prediction method based on H3 K27 mutation described in any one of claims 5-7.

Images