This article explains how the design and development of a research project conducted by Syracuse University’s Forensic and National Security Sciences Institute was informed by feedback from several criminal justice agencies and practitioners.
The feedback pertained to the interpretation of DNA samples that are a mix of material from multiple individuals. The first step in interpreting such a mixture is determining how many individuals contributed to the sample. The higher this number, the greater the difficulty of determining the number of individual contributors, a judgment that has relied on human interpretation. Under research funded by the National Institute of Justice (NIJ), the Syracuse University research team succeeded in developing a commercially available software called the Probabilistic Assessment for Contributor Estimation (PACE), which uses machine learning as a method of DNA mixture interpretation that runs on standard computers and estimates the number of individual contributors to a DNA sample at 98-percent accuracy in a few seconds. The keys to the successful transition of this technology to forensic practice relied on assistance in the process of licensing the PACE technology and the presence on the research team of a researcher who was previously a casework analyst. The latter factor helped to bring the perspective of the end user into the software development. The team also engaged development partners whose input added value to the features and uses of PACE.
Date Published: April 1, 2019