Deconvolution of DNA Mixtures Using Replicate Sampling and TrueAllele® Mixture Interpretation
Date
2020
Authors
Antillon, Sara
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Analysis of DNA mixture evidence does not always yield distinct profiles. This process is further complicated with low template DNA (LT-DNA)samples often seen in forensic casework. Traditional qualitative methods use thresholds to distinguish allele peaks from stutter peaks, noise, etc.resulting in data being omitted during analysis. In cases where LT-DNA is present, low peaks that could potentially be attributed to low contributor profiles may not be called due to these instituted thresholds. The probabilistic genotyping computer software program created by Cybergenetics (Pittsburgh, PA), TrueAllele® Casework, considers all data and performs quantitative analysis using probability to represent uncertainty. It objectively forms likelihood ratios (LR) that compare the probabilities of an evidentiary genotype with a suspect genotype relative to a reference population. A joint likelihood function (JLF) takes two or more independent sets of data and compares them jointly as opposed to a single event. The JLFcan elicit more identification information proving useful in DNA mixture analysis. This project used TrueAllele® Casework to perform DNA mixture analysis on two sets of previously published mixture data provided by Cybergenetics. The first set comprised 40 two contributor mixture samples and the second set included four sets of 10 randomized mixtures with two, three, four, and five contributors, respectively. The selected samples were interpreted singly and jointly in three variable groups: mixture weight, template concentration, and complex mixtures. The differences between the match logLRs of the single and joint analyses were calculated and an information gain was seen in all three groups when the samples were analyzed jointly. Changing DNA collection and amplification procedures for touch andDNA mixture evidence samples will increase the amount of data available forDNA mixture analysis using probabilistic genotyping. These procedures can be modified so that multiple swabs and replicate amplifications produce more data that TrueAllele can analyze using the JLF. Jointly analyzing each independent evidence data can lead to higher match statistics which will ultimately help in the identification of those who commit crimes.
Description
Keywords
DNA, Genotyping