Department of Forensic Molecular Biology, Erasmus University Medical Centre Rotterdam, Rotterdam, The Netherlands.
Forensic Sci Int Genet. 2011 Jun;5(3):170-80. doi: 10.1016/j.fsigen.2010.02.004. Epub 2010 Mar 27.
A new era of 'DNA intelligence' is arriving in forensic biology, due to the impending ability to predict externally visible characteristics (EVCs) from biological material such as those found at crime scenes. EVC prediction from forensic samples, or from body parts, is expected to help concentrate police investigations towards finding unknown individuals, at times when conventional DNA profiling fails to provide informative leads. Here we present a robust and sensitive tool, termed IrisPlex, for the accurate prediction of blue and brown eye colour from DNA in future forensic applications. We used the six currently most eye colour-informative single nucleotide polymorphisms (SNPs) that previously revealed prevalence-adjusted prediction accuracies of over 90% for blue and brown eye colour in 6168 Dutch Europeans. The single multiplex assay, based on SNaPshot chemistry and capillary electrophoresis, both widely used in forensic laboratories, displays high levels of genotyping sensitivity with complete profiles generated from as little as 31pg of DNA, approximately six human diploid cell equivalents. We also present a prediction model to correctly classify an individual's eye colour, via probability estimation solely based on DNA data, and illustrate the accuracy of the developed prediction test on 40 individuals from various geographic origins. Moreover, we obtained insights into the worldwide allele distribution of these six SNPs using the HGDP-CEPH samples of 51 populations. Eye colour prediction analyses from HGDP-CEPH samples provide evidence that the test and model presented here perform reliably without prior ancestry information, although future worldwide genotype and phenotype data shall confirm this notion. As our IrisPlex eye colour prediction test is capable of immediate implementation in forensic casework, it represents one of the first steps forward in the creation of a fully individualised EVC prediction system for future use in forensic DNA intelligence.
一个新的“DNA 智能”时代即将到来,这要归功于即将能够从生物材料(如犯罪现场发现的生物材料)中预测外部可见特征(EVC)。预计从法医样本或身体部位进行 EVC 预测,将有助于警方在常规 DNA 分析未能提供有价值线索的情况下,集中调查寻找未知个体。在这里,我们提出了一种强大而敏感的工具,称为 IrisPlex,用于在未来的法医应用中准确预测蓝色和棕色眼睛的颜色。我们使用了六个目前最能反映眼睛颜色的单核苷酸多态性(SNP),这些 SNP 之前在 6168 名荷兰欧洲人中显示了对蓝色和棕色眼睛颜色的预调预测准确率超过 90%。基于 SNaPshot 化学和毛细管电泳的单一多重分析,都广泛用于法医实验室,具有高水平的基因分型灵敏度,只需 31pg 左右的 DNA(大约六个人类二倍体细胞当量)即可生成完整的图谱。我们还提出了一种通过仅基于 DNA 数据进行概率估计来正确分类个体眼睛颜色的预测模型,并展示了在来自不同地理起源的 40 个人身上开发的预测测试的准确性。此外,我们使用 HGDP-CEPH 51 个人群的样本,了解了这六个 SNP 的全球等位基因分布。使用 HGDP-CEPH 样本进行的眼睛颜色预测分析表明,本研究中提出的测试和模型可以在没有先验祖先信息的情况下可靠地运行,尽管未来的全球基因型和表型数据将证实这一观点。由于我们的 IrisPlex 眼睛颜色预测测试能够立即应用于法医案件工作,因此它代表了在创建未来用于法医 DNA 智能的完全个性化 EVC 预测系统方面迈出的第一步。