Forensic DNA phenotyping (FDP) from low template DNA (LT-DNA) has advanced with the integration of machine learning (ML) models, improving predictive accuracy for traits like eye and hair color, with AUC > 0.9 for eye color and up to 15% improvement in SNP recovery. Tools such as HIrisPlex-S and VISAGE panels are effective for eye and hair color prediction, though challenges remain for skin tone, age, and facial morphology, especially in admixed populations. Future efforts should focus on multi-omics integration, portable sequencing, and inclusive datasets to enhance accuracy and ethical compliance in forensic applications.
5 citations
,
November 2020 in “Forensic Science International Genetics” Using trait prevalence priors in genetic prediction models for appearance traits is currently impractical due to limited knowledge and potential accuracy issues.
5 citations
,
September 2016 in “Security science and technology” DNA can predict physical traits like eye and hair color accurately, especially in Europeans, but predicting other traits and in diverse populations needs more research.
11 citations
,
February 2019 in “Research and reports in forensic medical science” DNA phenotyping helps predict physical traits from DNA with varying accuracy and requires careful ethical and legal handling.
June 2025 in “Albus Scientia” MC1R gene variations affect skin, hair color, UV sensitivity, and melanoma risk.