Allele frequencies and minor contributor match statistic convergence using simulated population replicates

Area of Science:

• Forensic Science
• Population Genetics
• Computational Biology

Background:

• Probabilistic genotyping compares forensic evidence by analyzing short tandem repeat (STR) alleles in mixed DNA profiles.
• Previous studies suggested non-contributors from low genetic diversity populations are more prone to misidentification in mixtures than those from high diversity populations.

Purpose of the Study:

• To investigate if allele distribution in a single reference population sample adequately represents genetic diversity for DNA mixture modeling.
• To determine the number of population replicates needed for reliable analysis using the forensim R package.

Main Methods:

• Simulated 200 US populations (50 each of African-American, Asian-American, European-American, and Hispanic descent) using the forensim R package.
• Generated and compared likelihood ratios for 2400 DNA mixtures against published data.
• Assessed the impact of population replicates on the reliability of results.

Main Results:

• Significant differences were observed when comparing simulated population data to published findings.
• A minimum of ten population replicates were necessary to minimize discrepancies with published data.
• Deviations from Hardy-Weinberg equilibrium and allele frequency distributions were noted in simulated populations.

Conclusions:

• The forensim R package is suitable for modeling population genetic parameters in forensic DNA analysis.
• A single population replicate, whether real or simulated, is insufficient for drawing definitive conclusions about DNA mixtures.
• Thorough evaluation of simulated populations for genetic parameters is crucial before employing them in DNA mixture modeling.