应用高通量测序DNA甲基化差异体液识别研究

The research of applying DNA methylation diversity analysis to forensically body fluids identification based on next-generation sequencing

目的通过检测和分析DNA甲基化差异的方法识别三种法庭科学犯罪现场常见体液(唾液斑、精液斑、血液斑)细胞来源。方法利用亚硫酸氢盐法对唾液斑、精液斑、血液斑的DNA样品进行甲基化处理,对由文献[1]和实验过程中筛选得到的甲基化位点进行illumina高通量测序,检测三种体液斑中细胞DNA的甲基化程度,并通过BP神经网络分析每一个位点的甲基化程度与细胞来源的关系及全部检测位点的甲基化程度联合与细胞来源的关系。结果运用BP神经网络对35份样品的73个位点的甲基化程度进行训练和预测,可以将唾液、精液、血液三种来源的DNA样本分开,其准确率达到77.3%。其中精细胞在L81528位点中表现为高甲基化,唾液和血液表现为低甲基化,可将精细胞准确鉴别出来。结论本研究建立的高通量测序分析DNA甲基化差异的方法可通过增加检测位点和样品数量进一步提高识别体液细胞来源的准确率。

Objective To identify three types of body fluids sources(saliva stains,semen stains and bloodstains from criminal cases based on the detection and analysis of DNA methylation diversity.Methods DNA samples of saliva stains,semen stains and bloodstains were methylated by bisulfite method,and the methylation loci selected from the literature[1]and the experiment were sequenced by illumina next-generation sequencing technology.The degree of DNA methylation of three types of body fluid strains was determined.Both the relationships between the degree of methylation of each locus and the cell source and the relationship between the degree of methylation of all detected loci and the cell source was analyzed based on BP neural network.Results This study could distinguish DNA samples from three types of sources(saliva,semen and blood),the accuracy rate was up to 77.3%,with the methylation degree of 73 loci from 35 samples detected and predicted by BP neural network analysis.L81528 locus in semen cells showed hypermethylation,but hypomethylation in saliva and blood,so the semen cells could be accurately identified.Conclusion The accuracy rate of the DNA methylation diversity analysis method based on next-generation sequencing technology could be further improved by raising the numbers of detected loci and samples.