The interpretation of low level DNA mixtures

The occurrence of mixed DNA profiles in forensic samples is not uncommon. Interpretation of these profiles however can be challenging. In this paper we compare three different models for interpreting mixed DNA profiles prior to the calculation of a statistical weight. Two of these models take into account the peak height of the alleles. The third method uses an unconstrained combinatorial approach. We compare the statistical weights calculated after applying the three different models to one low level two-person mixed DNA profile derived from a crime sample and provide tables of equations that can be applied to many different scenarios including single source and two and three person mixed DNA profiles.