Researcher says targeted microbial panel shows promise for identifying people from touch samples

Professor Bruce Fedolli of the Center for Human Identification at the University of North Texas Health Science Center presented pilot results showing that a targeted microbial panel can substantially increase sensitivity for human identification from skin swabs.

Fedolli told attendees the team developed a targeted multiplex, the HID SkinPlex, after finding that standard 16S rRNA profiling and shotgun metagenomics lacked the combination of depth and sensitivity needed for low‑level touch samples. “We may be getting better success with a targeted microbe set,” he said, explaining the approach uses PCR enrichment to amplify informative microbial gene targets.

The group first mined a longitudinal dataset (Oh et al.) that sampled 12 individuals at 17 body sites across three time points to identify markers that were both abundant and temporally stable. From those results, Fedolli said Propionibacterium acnes (p. acnes) emerged as a consistently abundant candidate; phylogenetic analyses using p. acnes showed individuals clustering across time points.

In a phase‑one pilot, the team swabbed eight individuals at three body sites (hand, foot, chest/manubrium) in triplicate (72 samples total). They tested roughly 286 microbial marker targets and designed about 572 primers for sequencing. Fedolli reported strong site‑specific results: hand samples reached accuracy near 98% in some analyses; manubrium results were similarly high in whole‑genome and targeted analyses; foot samples were more variable (mean ~73%, maximum ~83% in some cases).

As a proof of concept, Fedolli said his group also ran leftover DNA from the same swabs on a human forensic multiplex (an Illumina panel) and recovered roughly 30–52% of alleles from hand swabs, while the HID SkinPlex achieved near‑complete recovery for microbial targets in that comparison.

Fedolli emphasized limits and next steps. He said shotgun data are uneven across body sites and some taxa in the original dataset were absent from their pilot, which could reflect population differences or contamination in the source data. The team plans a larger validation study with 50 participants, will reduce amplicon sizes to help with degraded samples, and intends to move from presence/absence and nucleotide‑diversity metrics toward single‑nucleotide or sequence‑level markers and analytic weighting to convey evidence strength to fact‑finders.

He also highlighted open questions about environmental relatedness: “We have to look at people who are unrelated and related — unrelated are people who don't cohabitate and don't interact,” he said, noting studies must measure whether shared environments produce similarity that affects identification metrics.

Fedolli cited two papers documenting the work (one in Applied and Environmental Microbiology and a second in Forensic Science International: Genetics) and acknowledged graduate researcher Sarah Schmatis, Katherine Stevens and new group member Mike Coble. He said Schmatis has taken a position at the Centers for Disease Control and Prevention and thanked collaborators.

The presentation concluded with an invitation for questions; no formal actions or policy decisions were recorded in the session.