Incorporating Uncertainty Into Likelihood Ratios for DNA Evidence

• Activity Number: 313
• Type: Invited
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Committee of Representatives to AAAS
• Author(s): Bruce S. Weir*+ and John S. Buckleton and James M. Curran
• Companies: University of Washington and Environmental and Scientific Research and University of Auckland

Abstract

As DNA typing techniques become more sensitive, it becomes more likely that some alleles will "drop out" and not be detected. It is also possible that alleles may "drop in." Likelihood ratios work with the probabilities of the detected alleles in an evidence profile under alternative hypotheses about the contributors to the profile. If drop-out/drop-in are ignored these probabilities can be misleading: the probability will be zero if some alleles from a hypothesized contributor have dropped out, and additional contributors may be required to account for alleles that have dropped in. A common practice of ignoring loci that are not consistent with hypotheses because of undetected alleles or extra alleles may be prejudicial to a defendant. We report on efforts to include drop-in and drop-out rates in likelihood ratios, and we note the forthcoming recommendations tfor these inclusions from the International Society for Forensic Genetics.