Exploring DNA interpretation software using the PROVEDIt dataset

Abstract This work studies the effect of variability in the overall DNA typing process on the likelihood ratio (LR) generated from probabilistic genotyping (PG) systems. The data used in this study was obtained from the PROVEDIt set which is a comprehensive database openly available at lftdi.com [1]. The selected samples were composed of single-source and complex mixtures amplified with GlobalFiler kit (29 cycles) at different DNA template amounts and separated on 3500 Genetic Analyzer at different injection times. Mixtures were generated from different genotype combinations and contained varying contributor numbers and contributor ratios. Using fully continuous PG systems, we explore the effects of input DNA quantity, number of contributors, and mixture proportion, on the behaviour of the calculated LRs.