Evaluation of the Parabon® Snapshot™ DNA Phenotyping System

Purpose: DNA phenotyping, a relatively new area of forensic genetics, predicts a person’s ancestry and physical appearance (phenotype) from a DNA sample, by typing an array of single nucleotide polymorphisms (SNPs). The Parabon® Snapshot™ DNA Phenotyping System purports to predict detailed biogeographic ancestry (global and regional), sex, skin pigmentation, eye color, hair color, freckling, and face morphology in a single analysis. Methods: To assess the performance of this system, the University of North Texas Health Science Center recruited 24 individuals, representing phenotypic and ancestral diversity, to participate in a small blind study. Self-reported ancestry and phenotypes were collected and photographs were taken of each subject for subsequent comparison with the predictions generated. Twenty-five (25) DNA samples were sent anonymously to Parabon for Snapshot analysis. One sample was purposefully prepared as a mixture of two subjects. This sample manipulation was not made known to Parabon. Each DNA sample was genotyped and processed through the Snapshot prediction algorithms. All composites were generated at age 25 with a body mass index (BMI) of 22. Age and BMI targets were provided to Parabon subsequent to the blind analysis predictions, and composites for two subjects were blindly progressed by a Parabon forensic artist. Facial morphology predictions were compared to standardized craniofacial anthropologic measurements reported in the literature for each corresponding population. Results: The sex and bioancestry predictions for all participants were consistent with their self-reported classifications. Skin pigmentation predictions were relatively consistent within and among population groups for our sample set; 100% for Europeans (n=11) and 92% for non-Europeans (n=13). European eye and hair color predictions were 91% and 82% consistent, respectively. Conclusions: No conclusions were made regarding Snapshot’s performance for the prediction of facial morphology due to 1) the small sample size, 2) features used by Snapshot did not translate to those used by anthropologists, and 3) more anthropological data are needed. Although a small sample size, the results support that the majority of external feature phenotype and bioancestry predictions made with Snapshot are consistent with donor self-declaration and the Evaluators. A more thorough analysis is needed to assess Snapshot’s capacity to predict face morphology.