COMPLETION OF THE MIX 13 CASE STUDY BY EVALUATION OF MOCK MIXTURES WITH THE PROBABILISTIC GENOTYPING SOFTWARE GENOPROOF® MIXTURE 3
Maria Harthun, Anne-Marie Pflugbeil, Aline Jede, Holger Schönborn
Only recently, at the end of 2018, the results of the MIX 13 interlaboratory study, initiated by NIST in 2013, were published. Aiming to evaluate current available methods for DNA mixture analysis in the US and Canada, the complex DNA mixtures were reviewed from new perspectives, taking probabilistic genotyping methods into account. The presented scenarios were analysed with the software GenoProof® Mixture 3. For this purpose, the readily accessible sample data was used. The focus of the analysis was the evaluation of the results based on given references.
With the fully continuous approach of GenoProof® Mixture 3 it was possible to include or exclude the suspects of the cases correctly. LR calculations with the binary and semi continuous model can possibly lead to an over- or underestimation of the LR. The determination of the number of contributors also showed a solid result and corresponds to the possible solutions.
Increasing the contributors leads to an overestimation of the LR, caused by the increased number of allele combinations, which results in a weighting >0 if met in the simulation. In summary, a comparison of probabilistic software shows conclusive results for continuous models, but also indicates that these models have boundaries for the analysis of complex mixtures.