Back to Presentations
The first five exonerations using TrueAllele® statistical software: How labs can review and correct old cases
G. Hampikian and M. Perlin, "The first five exonerations using TrueAllele® statistical software: How labs can review and correct old cases", American Academy of Forensic Sciences 71th Annual Meeting, Baltimore, MD, 23-Feb-2019.
PowerPoint presentation of Dr. Hampikian and Dr. Perlin's talk.
TrueAllele® mixture interpretation software was used to analyze electronic STR data from crime scene samples in post-conviction claims of innocence. We describe the first five exonerations achieved using this DNA analysis software, and suggest improvements in crime lab processing that can help prevent wrongful convictions.
It has been well established through multiple studies, that complex DNA mixtures and low-template samples can result in conflicting interpretations and inconclusive results. These analysis problems can result in wrongful convictions. However, advanced statistical methods have been incorporated into software packages to better analyze complex DNA mixtures and low-template data. These computational approaches can be complemented by traditional manual data review.
In this study, we describe Cybergenetics’ use of TrueAllele analysis to examine post-conviction claims of innocence. This commercial “probabilistic genotyping” product uses a fully Bayesian probability model, solved by Markov chain Monte Carlo computer simulation. The presentation will describe five TrueAllele-based exonerations. These cases contain DNA mixture and low-template DNA evidence items, in cases where traditional methods were inconclusive. Cybergenetics automated TrueAllele re-evaluation of the electronic STR testing data was corroborated by manual scientist review.
We show that TrueAllele can be used to examine DNA evidence in old and current cases – even when only old electronic STR data is available, and no new DNA testing can be performed. In one case, two men were freed after 23 years in prison on dual life sentences for murder; Freddie Lawrence and Paul Jenkins were freed in Montana in 2018. Our computer analysis (corroborated by the state crime laboratory) produced a full STR profile that matched a new suspect in CODIS. In another case, two Indiana men who were wrongfully imprisoned for a total of 40 years were fully exonerated through TrueAllele computer analysis of “uninterpretable” low-level mixture data.
The described cases came from a partnership between the Boise State University laboratory of one author (GH), the Montana Innocence Project, the Georgia Innocence Project, the Illinois Innocence Project, the New Mexico Innocence Project and Pittsburgh-based Cybergenetics.
This research project was supported by grant award 2016-DY-BX-0006 from the Bureau of Justice Assistance. The Bureau of Justice Assistance is a component of the U.S. Department of Justice's Office of Justice Programs, which also includes the Bureau of Justice Statistics, the National Institute of Justice, the Office of Juvenile Justice and Delinquency Prevention, the Office for Victims of Crime, and the SMART Office.