Epigenetic clock calculation model and construction method

Abstract

The invention provides an epigenetic clock calculation model and a construction method, and belongs to the technical field of bioinformatics. A mathematical relationship between DNA methylation and age (month age) is established by taking northeast tiger as an example and taking excrement of the northeast tiger as a material, a method for constructing a calculation model agent of an epigenetic clock is provided, and the problem that in the prior art, materials needed for calculating the age of large animals (including wild animals) are not easy to obtain is solved. The calculation model of the epigenetic clocks is applied to calculation of the age of large animals or wild animals, selected materials are materials which are free of excrement damage and easy to obtain, and the accuracy of age calculation is higher than that of a calculation method using blood in a traditional method.

Description

Technical Field

[0001] The present invention belongs to the technical field of bioinformatics, and in particular relates to an epigenetic clock calculation model and a construction method. Background Art

[0002] Age is one of the basic parameters of a population and an important factor in animal ecology research. Age is of great significance in determining life history characteristics such as birth rate and mortality rate of animal populations. Age usually includes actual age and biological age. The actual age refers to the total time experienced since birth, while the biological age is the age inferred based on the anatomical and physiological developmental status, reflecting the actual state of the body's tissue structure and physiological function. The degree of conformity between biological age and actual age reflects the interaction relationship and interaction mechanism between the body and the environment, and also plays a key role in population monitoring, development trend prediction, life history research, and protection and management decision-making. Therefore, it is necessary to construct a method to accurately determine the age of species.

[0003] However, there is no age record for animals in the wild. Traditionally, the age of wild animals is estimated mainly based on appearance morphology, anatomical indicators (growth rings and wear of hard tissues, bone morphology measurements, size of food cuts, growth curves), physiological and biochemical indicators (sex hormone levels, chromosome telomere length, DNA methylation levels) and omics indicators (transcriptome, proteome, metabolome, methylome and metagenome). The appearance morphology observation method is often limited to animals with a long lifespan, whose body shape does not change much after adulthood, making it impossible to accurately determine their age. At present, the epigenetic clock is considered to be the most promising molecular method for estimating biological age. By establishing a model of the DNA methylation level of organism tissues and the age of the organism, the age of the animal can be predicted. The epigenetic clock has been applied to humans and experimental animals (such as mice), and the materials obtained include blood, muscle, saliva, oral swabs and semen. The prior art is recorded in Chronological Age Prediction: Developmental Evaluation of DNA Methylation-Based Machine Learning Models. Through the detection of blood samples, all R 2 The values ​​are all above 0.95. However, it is difficult to obtain the materials required by the prior art for some large animals (including wild animals). Therefore, the above method cannot complete the age estimation of some large animals (including wild animals), and other methods are also limited by the selection and acquisition of materials.

[0004] Usually, feces is the most easily accessible material for wild animals. The continuous division and differentiation of stem cells and the shedding of intestinal epithelial cells constitute a highly dynamic intestinal epithelium. Intestinal epithelial cells are one of the cells with the highest cell turnover rate (they are completely renewed approximately every 3 to 5 days). After completing their mission, these daughter cells fall off into the intestinal cavity and are excreted from the body with feces. The DNA obtained from feces mainly comes from the shed intestinal epithelial cells, and the huge number of intestinal epithelial cells in feces (the number of cells shed by mice every 24 hours can reach 2×10 8 , while humans can reach 10 11 The Zsolt Radák team found that increased entropy in the gut microbiome of physically active middle-aged / elderly people was associated with accelerated epigenetic aging, decreased health, or impaired health. These undoubtedly provide a good reference for solving the problem of animal age. However, there is currently no research on methods for estimating the age of large animals (including wild animals) through feces. Therefore, in order to estimate the age of large animals (including wild animals) more quickly, conveniently and accurately, a new age estimation method needs to be constructed. Summary of the invention

[0005] The present invention takes the Siberian tiger as an example and uses its feces as material to build a mathematical relationship between DNA methylation and age (age in months), thereby constructing a method for epigenetic clock calculation model, which solves the problem in the prior art that it is difficult to obtain the materials required for calculating the age of large animals (including wild animals).

[0006] A method for constructing an epigenetic clock calculation model, using feces as material to construct a model for calculating age through methylation sites of DNA extracted from feces.

[0007] The model is an elastic regression model, a support vector machine regression model or a random forest regression model.

[0008] The specific steps of the method are:

[0009] Step 1: Extract genomic DNA from feces;

[0010] Step 2: subjecting the genomic DNA to sulfite conversion treatment to obtain bisulfite-modified genomic DNA;

[0011] Step 3: construct a library of bisulfite-modified genomic DNA, then perform sequencing and data processing to obtain cytosine methylation reports and screen target methylation sites;

[0012] Step 4: Modeling based on target methylation sites and age.

[0013] The nuclear CT value of the genomic DNA in step 1 does not exceed 27.

[0014] The bacterial CT value of the genomic DNA described in step 1 is greater than 7.

[0015] The library construction kit used was xGen TM Methyl-Seq DNA Library Prep Kit.

[0016] The data processing is performed using the dplyr package, and according to the 3sigma principle, outliers are set to null values ​​and rows containing missing values ​​in the entire data set are deleted.

[0017] The screening condition of the target methylation site is that the correlation with age is greater than or equal to 0.5 or less than or equal to -0.5.

[0018] The number of the target methylation sites is 200, 400, 600 or 800.

[0019] A computational model of an epigenetic clock is constructed using the above method.

[0020] Beneficial Effects

[0021] Compared with the prior art, the method of constructing an epigenetic clock using fecal DNA in the present invention has the following advantages:

[0022] First, the selected material is feces, which is non-damaging and easily available. This solves the problem that when estimating the age of large animals (including wild animals) by traditional methods, the required materials including blood, muscle, saliva, oral swabs and semen are not easy to obtain. Therefore, the method constructed by the present invention has higher convenience and speed.

[0023] Secondly, a methodological process for screening age-related methylation detection sites based on fecal DNA self-built library sequencing and machine learning can be applied to the screening of age-related methylation difference sites in more animals, providing reference and inspiration for establishing age identification of other animals.

[0024] Finally, the method of constructing an epigenetic clock using fecal DNA in the present invention has a higher correlation than traditional blood testing, so the accuracy of the present invention is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The correlation between the predicted age and the actual age constructed by the four machine learning models for 400 methylation sites in the present invention. Figure 1 A represents the correlation of the elastic regression model, Figure 1 B represents the correlation of the artificial neural network regression model, Figure 1 C represents the correlation of the support vector machine regression model, Figure 1D represents the correlation of the random forest regression model. DETAILED DESCRIPTION

[0026] Example 1.

[0027] The present invention is further described below in conjunction with implementation examples, but the present invention is not limited to the following implementation examples.

[0028] Implementation condition 1: The nuclear CT value of the fecal DNA extracted by the present invention does not exceed 27, and the bacterial CT value is greater than 7;

[0029] Implementation condition 2: When the present invention uses stool DNA to build a library, the kit selected is xGen TM Methyl-SeqDNA Library Prep Kit;

[0030] Implementation condition three: The condition for the correlation between age and target methylation site in the present invention is greater than or equal to 0.5 or less than or equal to -0.5;

[0031] Implementation condition four: The present invention constructs a computational model of target methylation sites and age based on four models: elastic regression, support vector machine, artificial neural network and random forest regression, and determines the optimal parameters of each model and evaluates the effect through leave-one-out cross-validation.

[0032] The present invention is further described in detail below in conjunction with specific implementation modes.

[0033] 1. DNA extraction and DNA sulfite conversion treatment.

[0034] 1. Sample preparation.

[0035] The samples were feces of 33 captive Siberian tigers from the Siberian Tiger Park in Heilongjiang Province and Hengdaohezi. The samples were numbered and the birth date, sampling date, and gender of each individual were recorded, and the samples were stored in a -80℃ refrigerator.

[0036] 2. Extraction of sample DNA.

[0037] The genomic DNA was extracted from the above 33 stool samples according to the method established by our research group (patent announcement number: CN113186185B). The extracted DNA was labeled and stored at -20°C for future use.

[0038] 3. DNA sulfite conversion treatment.

[0039] The DNA extracted from 33 stool samples was subjected to sulfite conversion using EZ DNA Methylation-Gold Kit (D5006) (ZYMO, USA). The conversion was completed according to the instructions, and the resulting bisulfite-modified genomic DNA was stored in a -20°C refrigerator.

[0040] 2. Sequencing of self-built fecal DNA library and acquisition of methylated cytosine report for each sample.

[0041] Using xGen TM Methyl-Seq DNA Library Prep Kit constructs libraries for all bisulfite-modified genomic DNA. After the library is qualified (Qubit assay performs preliminary quantification of library concentration, and the total amount is greater than or equal to 200ng, it is qualified), high-throughput sequencing is performed on the Illumina platform, and sequencing is entrusted to Novogene. The sequenced data is aligned and deduplicated by Bsmap to obtain the cytosine methylation report of each sample.

[0042] 3. Data processing and screening of age-related methylation sites.

[0043] 1. Data processing.

[0044] The methylated cytosine reports of the 33 samples were integrated using the dplyr package (R language), outliers were detected according to the 3 sigma principle, outliers were set to null values, and rows containing missing values ​​in the entire data set were deleted.

[0045] 2. Screening of age-related methylation sites.

[0046] Based on Spearman correlation analysis, the correlation between the screened methylation sites and age was limited to less than or equal to -0.5 or greater than or equal to 0.5 to obtain the target methylation sites.

[0047] 4. Model, evaluate and predict target methylation sites and age.

[0048] 1. Determine the optimal number of target methylation sites for modeling.

[0049] Based on the nnet package, randomForest package, e1071 package, and glmnet package, different numbers (800, 600, 400, and 200) of target methylation sites were selected in descending order of relevance, and machine learning models of age and corresponding sites were constructed. The leave-one-out cross-validation method was used to evaluate each model, and the evaluation index of each model (R 2 ,MSE,RSME) and the optimal parameters of each model to select the number of optimal target sites involved in modeling, where R 2It represents correlation, MAE represents the mean of absolute errors (hereinafter referred to as errors), and RMSE represents the root mean square error. The evaluation criteria are that the higher the correlation, the smaller the error, and the higher the accuracy of the model.

[0050] 2. Determine the correlation between the predicted age of each model and the actual age.

[0051] After determining the optimal model and the optimal number of target sites involved in modeling, it is necessary to determine the correlation between the predicted age and the actual age of the four different models in order to screen the optimal model.

[0052] 5. Results

[0053] A total of 5840 target methylation sites with a correlation with age greater than or equal to 0.5 or less than or equal to -0.5 were screened out. In order to determine the optimal number of target methylation sites involved in modeling, elastic regression, support vector machine, artificial neural network and random forest regression calculation models with target methylation site numbers of 800, 600, 400 and 200 were constructed respectively. The results are shown in Table 1.

[0054] Table 1. Regression evaluation indicators for machine learning modeling with different numbers of target methylation sites.

[0055]

[0056] The results in Table 1 show that as the number of target methylation sites decreases, the accuracy of each regression model will increase, but when the number of sites decreases to 400, the accuracy of each model reaches the best state. Therefore, the optimal number of target methylation sites is 400.

[0057] To further verify the accuracy of the calculated age, we used the leave-one-out cross-validation method combined with four machine learning models: elastic regression, support vector machine, artificial neural network, and random forest to predict the age of 400 target methylation sites. The results are as follows Figure 1 ,The results showed that in addition to artificial neural network regression, ,elastic regression, support vector machine regression, and random forest regression models can ,all be applied to construct epigenetic clocks from fecal DNA.

[0058] According to the results in Table 1, when the number of target methylation sites is 400, when elastic regression is used to predict age, in the training set, the correlation between the predicted age and the actual age can reach 0.951, with an error of only 7.62 months, and in the test set, the correlation between the predicted age and the actual age can reach 0.984, with an error of only 9.15 months; when support vector machine regression is selected as the analysis model, in the training set, the correlation between the predicted age and the actual age can reach 0.993, with an error of only 3.89 months, and in the test set, the correlation between the predicted age and the actual age can reach 0. .972, with an error of only 7.50 months; when random forest regression was selected as the analysis model, in the training set, the correlation between the predicted age and the actual age could reach 0.960, with an error of only 7.70 months; in the test set, the correlation between the predicted age and the actual age could reach 0.970, with an error of only 10.42 months, further proving that support elastic regression, support vector machine regression, and random forest regression models can all be applied to constructing epigenetic clocks from fecal DNA, and that the support vector machine regression model is a superior model in age prediction based on DNA methylation.

Claims

1. A method for constructing an epigenetic clock computational model, characterized in that: The method uses feces as material and constructs a model for calculating age through methylation sites of DNA extracted from feces.

2. The method according to claim 1, characterized in that: The model is an elastic regression model, a support vector machine regression model or a random forest regression model.

3. The method according to claim 1, characterized in that: The specific steps of the method are: Step 1: Extract genomic DNA from feces; Step 2: subjecting the genomic DNA to sulfite conversion treatment to obtain bisulfite-modified genomic DNA; Step 3: construct a library of bisulfite-modified genomic DNA, then perform sequencing and data processing to obtain cytosine methylation reports and screen target methylation sites; Step 4: Modeling based on target methylation sites and age.

4. The method according to claim 3, characterized in that The nuclear CT value of the genomic DNA in step 1 does not exceed 27.

5. The method according to claim 3, characterized in that: The bacterial CT value of the genomic DNA described in step 1 is greater than 7.

6. The method according to claim 3, characterized in that The library construction kit used was xGen TM Methyl-Seq DNA Library Prep Kit.

7. The method according to claim 3, characterized in that The data processing is performed using the dplyr package, and according to the 3sigma principle, outliers are set to null values ​​and rows containing missing values ​​in the entire data set are deleted.

8. The method according to claim 3, characterized in that The screening condition of the target methylation site is that the correlation with age is greater than or equal to 0.5 or less than or equal to -0.

5.

9. The method according to claim 3, characterized in that: The number of the target methylation sites is 200, 400, 600 or 800.

10. A computational model of an epigenetic clock, characterized in that: The model is constructed using the method described in any one of claims 1-9.

Images