Forensic Science International: Genetics Volume: 9 Dated: March 2014 Pages: 102-110
This study tested a method for performing a useful direct search of DNA databases with a mixed DNA sample, using the framework of a likelihood ratio (LR) without the necessity of identifying a single contributor DNA profile.
Direct searching of unresolved DNA mixtures against DNA databases of known individuals is shown to be feasible as an investigative tool by using a suitable LR threshold that filters out high-probability false positives. Regardless of where the search LR threshold is set, there will always be the possibility of false positives (a LR threshold too low) and false negatives (a LR threshold too high). In two case examples, a LR of approximately 1 million resulted in links to individuals with high LR values. In recommending this proposed procedure, the authors caution that any positives selected should be used only for investigative purposes in setting priorities for more conclusive evidence collection. The empirical studies presented in this report focus on how to determine what constitutes a large LR. They also demonstrate the effect on a database search of complex mixed DNA profiles with contributors in equal proportions, with drop-out as a consideration. Attention is also given to the effect of an incorrect assignment of the number of contributors to a profile. The proposed method is tested with actual crime DNA samples that were previously considered unsuitable for DNA database comparisons. 3 figures, 7 tables, and 26 references
Report (Grant Sponsored)
Date Published: March 1, 2014