Forensic Human Identification Using Skin Microbiomes.

Area of Science:

• Microbiome Research
• Forensic Science
• Human Identification

Background:

• The human microbiome, unique to individuals, offers potential for forensic applications.
• Previous methods lacked stable skin microbiome features for long-term, high-accuracy identification.
• A need existed for robust methods to profile skin microbiomes for forensic human identification.

Purpose of the Study:

• To develop a novel approach for classifying skin microbiomes to their individual donors.
• To identify stable skin microbiome features for accurate, long-term human identification.
• To compare the efficacy of different feature types for microbiome classification.

Main Methods:

• Utilized Propionibacterium acnes pangenome presence/absence and nucleotide diversity of clade-specific markers.
• Employed supervised learning to classify skin microbiomes from 14 body sites across 12 individuals over >2.5 years.
• Applied feature selection to identify highly individualizing genetic markers.

Main Results:

• Achieved up to 100% classification accuracy for three body sites in attributing skin microbiomes to donors.
• Identified 187 stable genetic markers from 12 clades as highly individualizing.
• Nucleotide diversity features significantly outperformed presence/absence encodings for classification.

Conclusions:

• Supervised learning on skin microbiome features can accurately identify individuals over extended periods.
• Stable clade-specific markers and nucleotide diversity offer a promising foundation for forensic microbiome profiling.
• This approach advances the development of reproducible methods for forensic human identification using the skin microbiome.