Forensic DNA Phenotyping (FDP), which enables the prediction of externally visible characteristics (EVCs) and biogeographical ancestry from biological material, has emerged as a transformative yet ethically sensitive tool in modern criminal investigations. While FDP significantly aids in lead generation and identifying unknown persons by creating a "portrait" of the suspect without requiring a reference sample, its role as direct legal evidence remains highly debatable. This review critically assesses the scientific capacity and inherent limitations of FDP, alongside the ethical and legal challenges that currently impede its consistent and proportionate deployment across global jurisdictions. FDP models, demonstrate high predictive accuracy for distinct pigmentation traits (e.g., blue/brown eye color and dark skin categories). However, reliability diminishes significantly for intermediate phenotypes (e.g., intermediate eye color) and complex traits like facial morphology, which remain scientifically infeasible for forensic casework. Furthermore, for age estimation FDP relies increasingly on epigenetic markers; DNA methylation analysis, although its requirement for high-quantity DNA presents a significant constraint for trace evidence analysis. Despite its demonstrable utility as an investigative tool, FDP suffers from a profound global regulatory fragmentation, characterized by explicit authorization in some European nations contrasted with significant statutory gaps and contradictions in the United States. This lack of judicial clarity undermines the principle of proportionality and raises persistent ethical concerns regarding genetic privacy and the risk of institutionalizing racial bias. In conclusion, ensuring FDP legitimate use requires a concerted effort to establish validated, transparent, advanced methodologies, rigorous ethical guidelines, and judicial clarity, which are currently hampered by the lack of a unified regulatory framework, thus compromising fundamental fairness.
Abdelaal et al. (Sat,) studied this question.