What the study found
In 500 anonymized skeletal samples from unidentified human remains, the study found that common DNA quantification metrics were related to single nucleotide polymorphism (SNP) profile completeness, but not accurate enough to predict it reliably. The strongest signals came from measures reflecting the balance between human DNA and total DNA, including background DNA from non-human sources.
Why the authors say this matters
The authors conclude that, for forensic genome sequencing, current pre-sequencing metrics can help with some workflow decisions but are not sufficient predictors across the range of samples encountered in unidentified human remains. The study also states that MPS-based SNP profiling of unidentified human remains is highly effective and supports forensic genetic genealogy as the preferred approach for generating actionable genetic data for identification.
What the researchers tested
The researchers analyzed 500 anonymized skeletal samples submitted for forensic genome sequencing. They measured human-specific DNA using short and long autosomal quantitative PCR targets, total DNA using fluorometry, and compared these metrics with SNP call rate, which was used as a measure of profile completeness. They also examined bone type, degradation index, and machine-learning models for prediction.
What worked and what didn't
Of the 500 samples, 399 met the minimum human DNA threshold and were sequenced. Among sequenced samples, SNP call rates ranged from 8% to 91%, and 95.7% had call rates above 50%. The total:short DNA ratio and estimated human DNA input into library preparation showed the strongest correlations with call rate, while degradation index was only modestly associated; bone type affected whether samples advanced to sequencing, but call rate among sequenced samples was similar across major bone types. Machine-learning models reached moderate predictive performance, with the best validation R² at 0.47.
What to keep in mind
The abstract says the available DNA metrics are correlated with SNP profile completeness but are insufficient to predict it reliably for the sample range studied. The summary does not provide detailed limitations beyond the variability in DNA quality and quantity across bone samples.
Key points
- The study analyzed 500 anonymized skeletal samples from unidentified human remains.
- 399 samples met the minimum human DNA threshold and were sequenced.
- SNP call rates ranged from 8% to 91%, and 95.7% were above 50%.
- The strongest correlations with call rate involved the total:short DNA ratio and estimated human DNA input.
- Machine-learning prediction was only moderate, with a best validation R² of 0.47.
- Bone type influenced progression to sequencing, but not call rate among sequenced samples.
Disclosure
- Research title:
- DNA metrics only partly predicted SNP profile completeness
- Authors:
- Steven A Bates, Bruce Budowle, Morgan Johnson, Jianye Ge, Kristen Mittelman, David Mittelman
- Institutions:
- Biofuel Research Team, Opexa Therapeutics (United States), Opexa Therapeutics (United States), Opexa Therapeutics (United States), Opexa Therapeutics (United States), Radford University, Woodlands Hospital, Woodlands Hospital, Woodlands Hospital, Woodlands Hospital
- Publication date:
- 2026-02-25
- OpenAlex record:
- View
- Image credit:
- Bainscou, Wikimedia Commons, CC BY 3.0
Get the weekly research newsletter
Stay current with scholarly research without reading academic papers — one filtered digest, every Friday.