Prediction of people's origin from degraded DNA--presentation of SNP assays and calculation of probability

Int J Legal Med. 2013 Mar;127(2):347-57. doi: 10.1007/s00414-012-0728-0. Epub 2012 Aug 24.

Abstract

The characterization of externally visible traits by DNA analysis is already an important tool when investigating ancient skeletal remains and may gain similar importance in future forensic DNA analysis. This, however, depends on the different legal regulations in the different countries. Besides eye or hair color, the population origin can provide crucial information in criminal prosecution. In this study, we present the analysis of 16 single-nucleotide polymorphisms (SNPs) combined to two robust SNaPshot assays with a detection threshold of 25-pg DNA. This assay was applied to 891 people from seven different populations (West Africa, North Africa, Turkey, Near East, Balkan states, North Europe, and Japan) with a thorough statistical evaluation. The prediction model was validated by an additional 125 individuals predominantly with an ancestry from those same regions. The specificity of these SNPs for the prediction of all population origins is very high (>90 %), but the sensitivity varied greatly (more than 90 % for West Africa, but only 25 % for the Near East). We could identify West Africans with a certainty of 100 %, and people from North Africa, the Balkan states, or North Europe nearly with the same reliability while determination of Turks or people from the Near East was rather difficult. In conclusion, the two SNaPshot assays are a powerful and reliable tool for the identification of people with an ancestry in one of the above listed populations, even from degraded DNA.

Publication types

  • Validation Study

MeSH terms

  • DNA Degradation, Necrotic*
  • Electrophoresis
  • Ethnicity / genetics*
  • Genetics, Population*
  • Humans
  • Logistic Models
  • Models, Statistical
  • Multiplex Polymerase Chain Reaction
  • Polymorphism, Single Nucleotide*
  • Probability*
  • Racial Groups / genetics*
  • Reproducibility of Results
  • Sensitivity and Specificity