Effect of the Uncertainty in the Number of Contributors to Mixed DNA Profiles on Profile Interpretation

The study concludes that the assumption of an unreasonable number of contributors to a mixture being typed for DNA prior to interpretation may reduce the likelihood ratio (LR) of known incorrect inclusions. These are not desirable outcomes. The assumption of an increased number of contributors may significantly decrease the LR assigned to known minor and trace contributors to the mixture. This was observed in two key situations in this study. First, if there is no indication of an additional contributor, then the additional contributor's alleles must all be shared with the minor contributor, thus diffusing the seights across mnay more genotype combinations. Second, the treatment of a larger than threshold stutter as an additional contributor has almost no effect on the LR of the known contributors until that stutter peak is sufficiently large to interact with the minor contributors. When the additional height above expected stutter is similar in height to the alleles from a minor or trace component, the LR of that minor is decreased. STRmix does not currently contain allowance for forward stutter. Forward stutter occurs rarely and is typically removed at analysis prior to interpretation using some threshold based on empirical observations. Forward stutter incorrectly assigned as allelic may falsely imply an additional contributor or may confuse assignment of trace contributor alleles. The correct number of contributors in a profile is always unknown, but a reasonable assignment can often be made by counting alleles. 5 tables and 22 references