Optimizing a Human Forensic STR Panel for Non-Invasive Genetic Monitoring of Chimpanzees (Pan troglodytes)

Technical Summary: Optimizing a Human Forensic STR Panel for Non-Invasive Genetic Monitoring of Chimpanzees

Problem Statement
Reliable genetic identification is critical for conservation genetics, pedigree reconstruction, and wildlife forensics in endangered primates. In Cameroon, monitoring chimpanzees (Pan troglodytes) involves combining samples from wild populations and rescued captive individuals of uncertain ancestry. While non-invasive fecal sampling avoids animal capture, it frequently yields degraded, inhibitor-rich DNA that leads to amplification failure and genotyping errors. Although Single Nucleotide Polymorphism (SNP) approaches are emerging, microsatellites (Short Tandem Repeats or STRs) remain practical for individual identification due to their high allelic diversity and discriminatory power from few loci. However, while commercial human forensic STR systems have shown cross-amplification success in great apes using tissue samples, their performance on non-invasive fecal samples remains largely untested. There is a lack of standardized frameworks for optimizing these kits for degraded chimpanzee DNA.

Methodology
The study developed a standardized optimization framework to adapt the human forensic GlobalFiler™ PCR Amplification Kit (24 markers: 21 autosomal STRs, 1 Y-STR, 1 Y-indel, and Amelogenin) for chimpanzee genotyping.

• Sample Collection: The study utilized 5 blood samples and 77 fecal samples (27 from a sanctuary, 50 from the wild) in Cameroon.
• DNA Processing: DNA was extracted using specific kits for blood and stool. Endogenous chimpanzee DNA was quantified using the Quantifiler™ HP and Trio kits. Samples with ≥0.05 ng/μL DNA were selected for genotyping.
• Amplification and Genotyping: Samples were amplified in triplicate using the GlobalFiler™ kit. Fragment analysis was performed on a SeqStudio™ Genetic Analyzer.
• Data Curation:
  • Allele Binning: Since human nomenclature does not apply, alleles were defined by observed fragment lengths (bp), and locus-specific bins were manually established to accommodate microvariants.
  • Consensus Genotyping: A Quality Index approach required agreement between at least two of three replicates. Heterozygotes required a minor peak height of at least 50% of the major peak.
  • Filtering: Samples amplifying at fewer than 80% of loci were excluded to prevent poor sample quality from skewing locus performance metrics.
• Statistical Analysis:
  • Error Rates: Amplification Success (AS), Allelic Dropout (ADO), and False Allele (FA) rates were calculated.
  • Modeling: Beta regression models (logit link) were used to evaluate the effects of sample type and locus on AS, ADO, and FA.
  • Ranking: A composite scoring system was developed. Technical performance scores combined normalized AS (positive) and ADO/FA rates (penalties). This was further adjusted by the Probability of Identity among siblings (PID(sibs)) to create a final composite score reflecting both technical reliability and discriminatory power.

Key Results

• Locus Selection: From the original 24 markers, 5 were excluded due to failure to amplify, non-specific amplification, or low success rates (<50%). The remaining 16 autosomal loci formed the optimized panel.
• Performance Metrics: The 16-locus panel achieved a mean amplification success of 95.4%. Mean ADO was 0.08, and mean FA was 0.02.
  • Sample Type Variation: Blood samples showed negligible error. Sanctuary fecal samples had moderate errors, while wild field-collected fecal samples exhibited the highest error rates.
  • Locus Variation: Loci D10S1248, FGA, D2S441, and D3S1358 were identified as the highest-performing. Conversely, TH01, TPOX, CSF1PO, and D7S820 performed poorly.
• Individual Identification: The panel successfully identified 64 unique individuals (39 males, 25 females) from 69 high-quality samples. The cumulative probability of identity among siblings (PID(sibs)) was 1.2 × 10⁻⁶, indicating high discriminatory power.
• Sexing: Molecular sexing was successful for all 69 samples using the sex-linked markers.

Significance and Claims
The paper claims to provide a reproducible, standardized workflow for adapting commercial human forensic STR systems to non-invasive wildlife genetics.

• Framework Utility: The study demonstrates that the GlobalFiler™ kit can be effectively adapted for chimpanzee fecal DNA, moving beyond simple cross-species amplification to a quantitative, locus-level validation.
• Methodological Innovation: The authors highlight the development of a chimpanzee-specific allele binning framework and a composite locus-ranking system that integrates technical reliability (AS, ADO, FA) with discriminatory power (PID(sibs)). This approach prevents the misclassification of loci based solely on raw amplification patterns or small sample sizes.
• Transferability: The authors assert that this framework is easily transferable to other great ape species and wildlife conservation studies, offering a practical bridge between established microsatellite methods and emerging genomic approaches (SNPs, metabarcoding).
• Limitations: The authors modestly note that validation was restricted to Cameroonian populations, and broader geographic testing is required to refine locus rankings (e.g., TH01's low ranking may be population-specific). They also note that while the kit is effective, cost may limit large-scale ecological applications, suggesting the potential for developing smaller, chimpanzee-optimized panels based on their ranking data.