Probabilistic genotyping

Throughout the Probabilistic Genotyping of Evidentiary DNA Typing Results virtual workshop series, we have reviewed aspects of DNA mixture interpretation and have seen that probabilistic genotyping (PG) software can serve as a tool to assist the DNA Examiner in identifying possible genotype sets within a mixture and then calculating a likelihood ratio (LR). Approaches to modeling were shared by the developers of different PG software programs, followed by examples of internal validation studies and results.

Has a person of interest contributed DNA to a mixture obtained from evidence? Is that DNA associated with a crime? Both are questions that cannot be answered with absolute certainty. This module of the Probabilistic Genotyping of Evidentiary DNA Typing Results workshop series addresses the uncertainty and limitations of probabilistic genotyping systems that are used to calculate likelihood ratios and infer genotype sets from DNA typing results.

Biological modeling, statistical theory, and the computer processing of probabilistic genotyping systems are based on long- standing principles, yet their usage in U.S. courts has been limited to the past few years. This module of the Probabilistic Genotyping of Evidentiary DNA Typing Results web series addresses topics relevant to the admissibility of probabilistic genotyping results and the presentation of results in criminal proceedings. Various decisions related to probabilistic genotyping will be summarized from a legal perspective.

Historically, the predominant means of assigning statistical weight to many DNA mixtures in the U.S. has been the combined probability of inclusion, or CPI. Accordingly, many forensic DNA analysts, as well as attorneys and judges, are well-versed in the CPI as a match probability. In the fifth module of this workshop series devoted to probabilistic genotyping, the presenters will help DNA analysts and legal professionals understand degrees of statistical weight and properly communicate likelihood ratios and their meaning.

The Scientific Working Group on DNA Analysis Methods (SWGDAM) and other bodies have published guidelines that specify empirical studies to be performed to assess reliability, identify limitations and develop suitable usage conditions for probabilistic genotyping systems.

Over a dozen probabilistic genotyping software programs are commercially available or accessible as freeware. This module presents an instructive overview by software developers of three programs that are supported by published developmental validation.

This webinar is part 3 of the Probabilistic Genotyping Of Evidentiary DNA Typing Results Webinar Series and was presented by the NIJ Forensic Technology Center of Excellence on May 22, 2019.

Probabilistic genotyping is a tool that uses computing power to aid in the identification of possible genotype sets within DNA typing results and to calculate likelihood ratios to estimate evidentiary weight. In this installment of Probabilistic Genotyping of Evidentiary DNA Typing Results, we will detail the background and principles of biostatistical analysis, to include match probabilities, likelihood ratios and other specific topics aimed at furthering understanding of the statistical basis of probabilistic genotyping.

Based on the review of various forms of STR typing results and factors that impact the presence or condition of DNA on evidence, instructors in this installment of Probabilistic Genotyping of Evidentiary DNA Typing Results lay a foundation for the interpretation of forensic DNA typing results. The session will begin with background information on forensic DNA analysis, including terminology and mechanisms, to help all participants comprehend the material covered in this series.