A classifier for the SNP-based inference of ancestry.

摘要: Ancestral inference from DNA could serve as an important adjunct for both standard and future human identity testing procedures. However, current STR methods the of ancestral affiliation have inherent statistical technical limitations. In effort to identify bi-allelic markers that can be used infer DNA, we screened 211 SNPs in pigmentation xenobiotic metabolism genes. Allele frequencies 56 (most genes) were dramatically different between groups unrelated individuals Asian, African, European descent, observed simulated log likelihood ratios revealed exceptional value inference. Log multilocus estimates biological ancestry (EAE/EBA) ranged 7 10, which are on par with best batteries yet described. A linear classification method was developed incorporating these into a classifier model 99, 98, 100% accurate identifying European, Asian respectively. The describe therefore first step development practical multiplex test forensics setting.