长度多态与序列多态STR模型法庭科学参数比较

摘要

基于毛细管电泳平台进行STR基因座分型是当前个体识别的金标准.二代测序技术支持STR序列多态分型,并有可能在法庭科学领域被广泛应用.相比长度多态性,STR测序可提供更大的信息量,但相关法庭科学参数的定量计算十分必要.本文建立了简单的STR基因座模型,分别计算了长度多态和序列多态STR模型基因座的法医遗传学参数,结果表明:对于单个STR基因座,其序列多态模型的个体识别力和非父排除率相比长度多态模型更高,说明序列多态STR遗传标记识别无关个体及排除非父的能力更强.在联用15个非连锁遗传模拟基因座进行法医DNA分析时,长度多态模型和序列多态模型的累积匹配概率分别在10-18和10-26量级;而如果要达到长度多态模型15个基因座的累积匹配概率(10-18),仅需使用10个非连锁模拟序列多态STR基因座.希冀此模型比较能为二代测序STR数据的法庭科学应用提供参考.%Capillary electrophoresis-based STR genotyping is accepted as the gold standard for human individual identification. Next generation sequencing (NGS) allows for the full resolution of STR base composition, and has the potential to be widely used in the field of forensics. Compared with length polymorphism, STR sequencing could provide more information, and quantitatively calculating the forensic parameters is necessary. In this paper, we established simple models for length-based and sequence-based STRs, and calculated the forensic parameters for both models. The results showed that for a single STR locus, compared with length polymorphism, STR sequence polymorphism could provide higher power of discrimination and power of exclusion, indicating sequence-based STR marker have stronger ability for identifying unrelated individuals and exclude non biological father. By combining 15 non-linkage loci for forensic DNA analysis, the cumulative matching probability values for length-based and sequence-based STR models are at 10-18and 10-26levels, respectively. Only 10 non-linkage sequence-based STR is required to reach a cumulative matching probability of as high as 15 length-based STR loci. It is hoped that these simulated models and calculations can provide a reference for the forensic application of NGS-based STR genotyping.