Advances in DNA Phenotyping: From Molecular Markers to Predictive Forensic Profiling

Abstract: Forensic DNA phenotyping (FDP) represents a revolutionary advance in human identification, allowing the prediction ofexternally visible characteristics (EVCs) such as eye, hair, and skin color directly from genetic material. Unlike short tandem repeat(STR) profiling, which requires comparison to known DNA references, FDP enables phenotype inference when no match exists inDNA databases. This systematic review synthesizes developments from 2020–2025, focusing on molecular markers, genomicsequencing technologies, computational algorithms, and ethical frameworks. Following PRISMA guidelines, relevant peer-reviewed literature was screened across PubMed, Scopus, and Web of Science. A total of 72 studies met inclusion criteria. The findings reveal rapid diversification from pigmentation-associated SNPs toward multi-omics approaches integrating genomic, epigenetic, and transcriptomic markers. Advances in machine learning—particularly deep neural networks and random forests—have enhanced prediction accuracy, especially when ancestry information is integrated. However, substantial variability remains in facial morphology prediction and population transferability. Ethical challenges persist regarding privacy, consent, and potential misuse of phenotypic data. The review concludes that FDP is transitioning from experimental to operational forensic science, yet its societal legitimacy will depend on transparent validation, equitable access, and regulatory oversight.