Internal validation of STRmix™ – A multi laboratory response to PCAST

Abstract

We report a large compilation of the internal validations of the probabilistic genotyping software STRmix™. Thirty one laboratories contributed data resulting in 2825 mixtures comprising three to six donors and a wide range of multiplex, equipment, mixture proportions and templates. Previously reported trends in the LR were confirmed including less discriminatory LRs occurring both for donors and non-donors at low template (for the donor in question) and at high contributor number. We were unable to isolate an effect of allelic sharing. Any apparent effect appears to be largely confounded with increased contributor number.

The Probabilistic Genotyping Software STRmix: Utility and Evidence for its Validity

“…13 of Bright et al. ) shows the level of over and underestimation of the apparent NoC ( N ) determined following the individual laboratories’ protocols compared to the target N in their respective studies. Overestimation of N generally led to similar or lower LR s for true contributors.…”

“…Recently, the internal validation data from 31 laboratories using or validating STRmix™ were compiled and interpreted (hereafter “internal compilation study” ) specifically to address the points raised within the PCAST Report. This study concluded that this combined dataset “demonstrates a foundational validity of, at least, the STRmix™ software method for complex, mixed DNA profiles to levels well beyond the complexity and contribution levels suggested by PCAST.” These efforts, representing a substantial resource commitment, are a collation of the validation studies from 31 laboratories and demonstrate that there is support for interpreting a minor contributor much less than 20%, and in fact down to 0% (present but not observed), of the total DNA present in the mixture.…”

“…Importance sampling creates a biased sample by drawing from a distribution of importance, in the DNA example, these are profiles likely to produce high LR s. As the sample is biased, it is necessary to readjust for the bias after simulation. In a large set of mixtures compiled from 31 laboratories , all large (over 10,000) LR s for nondonors were investigated; in all instances, the nondonors had high allelic overlap with the profile. This is a correct result.…”

“…However, one can have confidence that a breadth of normal usage is well tested, as exemplified by Bright et al. with more than 2,825 mixtures of DNA from three to six contributors.…”

“…The internal validation compilation study described this effect of estimating various NoC to a mixture. Figure (derived from fig.…”