Internal validation of probabilistic genotyping software STRmix™ V2.8 for GlobalFiler™ profiles generated from Japanese individuals

The interpretation of mixed DNA profiles originating from multiple individuals can be very challenging. Probabilistic genotyping software is used to interpret such mixed DNA profiles, and it assigns a likelihood ratio based on the evidence given two hypotheses: a prosecution hypothesis and a defense hypothesis. For proper utilization of probabilistic genotyping software, internal validation is required. In this study, we performed internal validation of the probabilistic genotyping software STRmix™. Our internal validation studies complied with the Scientific Working Group on DNA Analysis Methods guidelines for the validation of probabilistic genotyping systems, and was performed using GlobalFiler™ profiles generated from Japanese individuals to study the sensitivity, specificity, and precision of STRmix™, and the effects of the addition of a known contributor and incorrect assumption of the number of contributors were also assessed. The results of our internal validation studies suggest that STRmix™ using our laboratory-specific parameters is suitable for the purpose of interpreting mixed DNA profiles and statistical evaluation in our experimental and analytical environment. In addition, there were a few rare cases in which STRmix™ interpreted the results as exclusion (likelihood ratio = 0), despite the person of interest being a true contributor, due to extreme heterozygote imbalance and/or significant differences in the mixture ratio between loci due to the stochastic effects of PCR amplification. Therefore, we report these rare cases along with the results of our internal validation studies. • The first internal validation of STRmix™ using Japanese individuals in forensics. • The analysis was performed according to the SWGDAM validation guidelines. • STRmix™ is suitable for interpreting Japanese mixed DNA profiles.