Methylation characteristic data detection method and detection kit

Abstract

The invention belongs to the technical field of biomedicine, and relates to a methylation characteristic data detection method and a detection kit. The method comprises the following steps: extracting genome DNA from a tissue sample; performing high-throughput detection on the methylation level of the CpG site in the genome DNA to obtain a detection result; constructing a methylation characteristic analysis model based on the methylation state of the CpG site in the detection result; generating methylation feature data of the tissue sample through the model; the methylation characteristic data is used for detecting a kidney cancer specific DNA methylation marker associated with the abnormal tricarboxylic acid cycle metabolism. Through integration of multi-omics data and strict screening standards, detection of needle biopsy and surgical resection samples is supported, compatibility is high, generation and analysis of methylation data are focused, and disease diagnosis or treatment is avoided.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a methylation characteristic data detection method and a detection kit. Background Art

[0002] Renal cell carcinoma (RCC) is a common malignant renal tumor. In addition to the common types, there are also special, rare subtypes, such as fumarate hydratase-deficient renal cell carcinoma (FH-RCC) and succinate dehydrogenase-deficient renal cell carcinoma (SDH-RCC). These special subtypes, characterized by tricarboxylic acid cycle defects, are more invasive and metastatic, with clinical manifestations including a high rate of metastasis at initial diagnosis, a high rate of early postoperative recurrence, and poor efficacy of systemic treatment. Currently, the diagnosis of this type of renal cell carcinoma relies on molecular pathology and genetic diagnosis of postoperative specimens. For patients with advanced disease, there is a lack of standard treatment options and effective efficacy prediction methods.

[0003] Imaging is an important diagnostic method, but the imaging manifestations of tricarboxylic acid cycle defect-related renal cancer are atypical, making it difficult to detect preoperatively through imaging methods. Although postoperative pathological diagnosis can assist in diagnosis, the significant morphological heterogeneity of FH-RCC and SDH-RCC tumor cells leads to very high rates of missed diagnosis and misdiagnosis in cases without clinical suspicion or family history. In addition, tumor genetic testing is expensive, and FH and SDHx gene mutations are highly random, with no clear high-frequency mutation hotspots. The testing methods and standards of different laboratories also vary, making it challenging for clinicians to interpret the results.

[0004] In view of this, this application is hereby filed. Summary of the Invention

[0005] The purpose of the present invention is to provide a methylation signature data detection method and detection kit. Based on multi-omics integrated analysis and the independently established largest known sample library of FH-RCC and SDH-RCC in China and abroad, the method analyzes the CpG island methylation phenotypic characteristics of tricarboxylic acid cycle deficiency-related renal cancer and identifies a series of key CpG island methylation sites with diagnostic and prognostic value.

[0006] The present invention is achieved through the following technical solutions:

[0007] In a first aspect, a method for detecting methylation signature data is proposed, comprising the following steps: extracting genomic DNA from a tissue sample; performing high-throughput detection on the methylation level of CpG sites in the genomic DNA to obtain a detection result; constructing a methylation signature analysis model based on the methylation status of the CpG sites in the detection result; generating methylation signature data of the tissue sample through the model; and using the methylation signature data to detect renal cancer-specific DNA methylation markers associated with abnormal tricarboxylic acid cycle metabolism.

[0008] Furthermore, the high-throughput detection method is: using the Infinium Methylation EPIC Bead Chip chip to detect the methylation status of 450,000 CpG sites in the DNA sample to obtain genome-wide methylation data.

[0009] Furthermore, the construction of the methylation feature analysis model includes the following steps: screening differentially methylated CpG sites through differential methylation analysis; dividing the samples into two categories: CpG island methylator phenotype (CIMP) and non-CIMP based on unsupervised cluster analysis; and screening CpG sites related to the tricarboxylic acid cycle metabolic pathway through machine learning methods.

[0010] Furthermore, the method for screening out differentially methylated CpG sites is as follows: screening out differentially methylated CpG sites with FDR < 0.05 and methylation degree difference β > 0.3.

[0011] Furthermore, the machine learning method includes one or more of random forest, support vector machine or LASSO regression.

[0012] In a second aspect, a detection kit for methylation characteristic data is proposed, comprising probes or primers, wherein the probes and primers are used to detect the methylation status of the CpG sites and extract DNA.

[0013] Furthermore, the kit also includes data analysis and model building.

[0014] Compared with the existing technology, the present invention has the following advantages and benefits: 1. Through the integration of multi-omics data and strict screening criteria, the model achieved an AUC of 0.99 in an independent validation cohort; 2. The kit supports the detection of puncture biopsy and surgical resection samples and has strong compatibility; 3. It focuses on the generation and analysis of methylation data, avoiding involvement in disease diagnosis or treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without making any creative efforts.

[0016] Figure 1 A schematic flow chart of a method for detecting methylation characteristic data provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0018] Example

[0019] like Figure 1 As shown, a method for detecting methylation characteristic data provided by a first aspect of an embodiment of the present invention includes the following steps:

[0020] Step 1: Extract genomic DNA from tissue samples.

[0021] Specifically, high-quality genomic DNA was extracted from patient tissue samples (tumor or paired normal tissue), ensuring a tumor cell ratio of at least 70%. The sample size was 1,200 cases.

[0022] Step 2: Perform high-throughput detection on the methylation level of the CpG sites in the genomic DNA to obtain detection results.

[0023] Specifically, the Infinium MethylationEPIC BeadChip chip was used to detect the methylation status of 450,000 CpG sites in DNA samples to obtain genome-wide methylation data.

[0024] Step 3: Based on the methylation status of the CpG sites in the detection results, a methylation feature analysis model is constructed.

[0025] Comparing tumor samples with normal controls, the MethylKit+ChAMP analysis process was used to screen for significantly differentially methylated CpG sites, specifically hypermethylated and hypomethylated sites associated with the development and progression of Krebs-related renal cancer. Conventional criteria for differentially methylated sites are FDR < 0.05 and methylation level difference β > 0.05. Based on the project team's previous experience, the differential criteria for this tumor type can be more stringent, with FDR < 0.05 and methylation level difference β > 0.3.

[0026] Step 4: Generate methylation feature data of the tissue sample using the model.

[0027] In the past, conventional methylation models only considered differential methylation sites. This embodiment is based on unsupervised cluster analysis of differential methylation sites, and divides the samples of renal cancer related to tricarboxylic acid cycle defects into two categories: CIMP and non-CIMP, and observes the significant association between CIMP and tumor malignant behavior, including high TMB, chromosomal abnormalities and PD-L1 positivity. From the whole genome methylation data, combined with a variety of machine learning methods (including random forest, support vector machine, LASSO regression, etc.), candidate methylation sites are screened out. Through multi-omics data association analysis, key sites associated with multi-omics features of renal cancer related to the tricarboxylic acid cycle are further identified. Finally, a combination model of 4 specific CpG sites was identified, which can efficiently and accurately distinguish between a variety of molecular subtypes of renal cell carcinoma. Specifically, from the whole genome methylation data, candidate methylation sites are screened out by machine learning methods (including random forest, support vector machine, LASSO regression, etc.). Based on a multi-omics database and case cohort of tricarboxylic acid cycle defect-related renal cancer, the genomic, transcriptomic, and metabolomic data of tricarboxylic acid cycle-related renal cancer were integrated for association analysis. Key sites associated with the multi-omics characteristics of tricarboxylic acid cycle-related renal cancer were further identified through multimodal, multi-omics models. The selected sites were associated with the mutation characteristics of tricarboxylic acid cycle defect genes such as FH and SDH (P<0.05), and the absolute value of the correlation with the expression level of this group of genes was greater than 0.4, or the absolute value of the correlation with the metabolite concentration was greater than 0.4. Finally, a combined model of 4 specific CpG sites was identified. This model can efficiently and accurately distinguish various molecular subtypes of renal cell carcinoma (the test set AUC is 1, and there is no previous report on the diagnostic efficacy of similar models), as well as the tissue origins of other urinary system tumors.

[0028] The diagnostic performance of the above model was evaluated in an independent external validation cohort, and the results showed that this CpG site combination had significantly high sensitivity and high specificity in the validation set (AUC of 0.99) in distinguishing tricarboxylic acid cycle deficiency-related renal cancer and other urinary system tumors.

[0029] A second aspect of this embodiment provides a kit for detecting methylation signature data. The kit includes probes or primers for detecting the methylation status of CpG sites, as well as auxiliary reagents for DNA extraction, amplification, and detection. The kit also includes software tools for data analysis and model building, suitable for rapid diagnosis and personalized prognostic management of renal cancer associated with tricarboxylic acid cycle defects in clinical practice.

[0030] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting methylation characteristic data, characterized in that: The following steps are involved: Extract genomic DNA from tissue samples; Performing high-throughput detection on the methylation level of the CpG site in the genomic DNA to obtain a detection result; Based on the methylation status of the CpG sites in the detection results, a methylation feature analysis model is constructed; The model is used to generate methylation feature data of the tissue sample; the methylation feature data is used to detect renal cancer-specific DNA methylation markers associated with abnormal tricarboxylic acid cycle metabolism.

2. A method for detecting methylation characteristic data according to claim 1, characterized in that: The high-throughput detection method is: using the Infinium MethylationEPIC BeadChip chip to detect the methylation status of 450,000 CpG sites in a DNA sample to obtain genome-wide methylation data.

3. The method for detecting methylation characteristic data according to claim 1, wherein: The construction of the methylation feature analysis model comprises the following steps: Differentially methylated CpG sites were screened through differential methylation analysis; Based on unsupervised cluster analysis, the samples were divided into two categories: CpG island methylator phenotype (CIMP) and non-CIMP; CpG sites related to the tricarboxylic acid cycle metabolic pathway were screened using machine learning methods.

4. A methylation feature data detection method according to claim 3, characterized in that: The method for screening differentially methylated CpG sites is: screening differentially methylated CpG sites with FDR < 0.05 and methylation degree difference β > 0.

3.

5. The method according to claim 3 or 4, characterized in that The machine learning method includes one or more of random forest, support vector machine or LASSO regression.

6. A kit for detecting methylation characteristic data, characterized in that: It comprises probes or primers, and the probes and primers are used to detect the methylation status of the CpG sites and extract DNA.

7. The detection kit according to claim 6, characterized in that The kit also includes tools for data analysis and model building.

Images