Introduction

Success Story: Improving DNA Mixture Interpretation with the Help of Machine Learning

Success Story: Improving DNA Mixture Interpretation with the Help of Machine Learning

National Institute of Justice and Syracuse University

Date

March 2019

Overview

When DNA is recovered from a crime scene, victim, or suspect, it may be a mixture of genetic information from multiple individuals. To successfully interpret a mixture of DNA, analysts must first determine the number of contributors present in the sample. This is a critical but challenging first step towards “deconvolution” of the DNA data. Standard computer programs designed to assist with DNA mixture analysis require analysts to determine the number of contributors beforehand. To do so, users must identify, categorize, and remove irrelevant data called artifacts. However, in cases when the DNA is in limited quantities, degraded, or both—individual contributors are difficult to separate, some even artifacts. Generally, as the number of contributors increases and complexity of the DNA profile increases, requiring greater amounts of user time and experiential judgment to determine the number of contributors. To address these challenges, Dr. Michael Marciano and Jonathan Adelman of Syracuse University developed the Probabilistic Assessment for Contributor Estimation (PACETM) a machine learning method to improve mixture interpretation. PACETM is unique in the field of forensic DNA analyses and is leading the way toward a future filled with myriad new tools and interpretation methods that better utilize complex, challenging samples. The development of PACETM (patent pending), which is exclusively licensed to NicheVision, was supported through the National Institute of Justice (NIJ) forensic science R&D program with the goal of eventually seeing operational use in casework.

“This is something that every DNA crime lab worldwide should investigate. We now have a tool that with the press of one button does the entire suite of services and provides you with the end result. Having the simple decisions made for you enables focus on addressing the issues of most importance.”

- Vic Meles | Marketing & Business Director, NicheVision



Funding for this Forensic Technology Center of Excellence success story was provided by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice.

The opinions, findings, and conclusions or recommendations expressed in this success story are those of the author(s) and do not necessarily reflect those of the U.S. Department of Justice.

Contact us at ForensicCOE@rti.org with any questions and subscribe to our newsletter for notifications.


Related Content

Hair Root Staining – What Can Hematoxylin Do for Your Laboratory?

This webinar originally occurred on August 16, 2022 Duration: 1 hour Overview Hair evidence collected as part of a forensic investigation has the potential to provide valuable source information through DNA analysis of its root. In the years prior to…

Interlaboratory Comparison of SpermX and Conventional Differential Extractions

This webinar originally occurred on Thursday, August 11, 2022 Duration: 1 hour Overview Scientists from three participating laboratories – University of Central Florida (UCF), Center for Forensic Science Research and Education (CFSRE), and InnoGenomics (IGT) – presented results obtained from…