Suppr超能文献

香农准则在法医Y染色体短串联重复序列(Y-STR)标记选择中的应用

Shannon's equivocation for forensic Y-STR marker selection.

作者信息

Siegert Sabine, Roewer Lutz, Nothnagel Michael

机构信息

Department of Statistical Genetics and Bioinformatics, Cologne Center for Genomics, University of Cologne, Weyertal 115b, 50931 Cologne, Germany.

Dept. Forensic Genetics, Institute of Legal Medicine and Forensic Sciences, Charité - Universitätsmedizin Berlin, 13353 Berlin, Germany.

出版信息

Forensic Sci Int Genet. 2015 May;16:216-225. doi: 10.1016/j.fsigen.2015.02.001. Epub 2015 Feb 9.

Abstract

Short tandem repeat (STR) markers are widely and continuously used in forensic applications. However, past research has demonstrated substantial allelic association between STR markers on both autosomes and the X chromosome, leading to partially redundant information that these markers can provide. Here, we quantify the allelic association between Y-chromosomal STR markers that are part of established forensic panels, separately for three different continental groups. We further propose a sequential marker selection procedure that is based on Shannon's equivocation and that accounts for allelic association between STR markers, leading to a maximal gain in independent information. In application to three real-world data sets, we demonstrate the procedure's superior performance when compared to single-locus diversity selection strategies, resulting in the optimal marker set for a given data set in the majority of marker subsets. Noting the inferior performance of the established Y-STR marker panels in a retrospective investigation, we suggest that future forensic marker selection should be guided, besides by other technical selection criteria, by an equivocation-based approach to obtain maximally discriminatory marker sets at minimal cost.

摘要

短串联重复序列(STR)标记在法医鉴定中得到广泛且持续的应用。然而,过去的研究表明,常染色体和X染色体上的STR标记之间存在显著的等位基因关联,导致这些标记所能提供的信息存在部分冗余。在此,我们分别针对三个不同的大陆群体,对作为既定法医鉴定面板一部分的Y染色体STR标记之间的等位基因关联进行了量化。我们进一步提出了一种基于香农疑义度的序列标记选择程序,该程序考虑了STR标记之间的等位基因关联,从而实现独立信息的最大增益。在应用于三个实际数据集时,我们证明了该程序与单基因座多样性选择策略相比具有更优的性能,在大多数标记子集中为给定数据集生成了最优标记集。鉴于既定的Y-STR标记面板在回顾性调查中的表现不佳,我们建议,除了其他技术选择标准外,未来的法医标记选择应以基于疑义度的方法为指导,以最小的成本获得最大区分性的标记集。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验