MODULE 2: STATISTICAL GENETICS AND THE MECHANISMS OF PROBABILISTIC GENOTYPING

← Back to Series 

This module originally occurred on May 8,  2019
Duration: 4 hours

Overview

Module 2: Statistical Genetics and the Mechanisms of Probabilistic Genotyping

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.

To begin, we will introduce Fst (sometimes called theta), the population genetics parameter that measures remote relatedness between apparently unrelated individuals.  We will then derive single-locus match probability formulas based on Fst in a simple model of population genetics.  Next, we consider appropriate values for Fst, and estimates of allele sampling probabilities based on database counts.  The validity of multiplying match probabilities across loci, sometimes called the "product rule," will be discussed.

We’ll touch on some more complex issues: relatedness, mixed and low-template profiles, and the connection between match probabilities and likelihood ratios. To calculate a likelihood ratio, the analyst must develop two propositions; in simplest form – for a DNA result originating from one individual – one would consider that “the DNA is from the person of interest” and “the DNA is from an unknown, unrelated individual.” Using a variety of case scenarios ranging from simple to complex, the strategy of devising propositions and dealing with uncertainty in the number of contributors to DNA mixtures will be detailed, along with the resulting impact on the likelihood ratio. Participants will be guided through practical exercises in determining the number of contributors, developing propositions and calculating the likelihood ratio.

Detailed Learning Objectives

1. Articulate the theory and application of statistical techniques used in probabilistic genotyping, considering both unrelated and related individuals, and perform calculations relevant for matching DNA profiles.
2. Understand the reasoning behind the Balding-Nichols match probabilities, including its limitations.
3. Be able to assess appropriate values for Fst given the circumstances of a particular case.
4. Appreciate the different analyses appropriate for autosomal and unilineal DNA profiles.
5. Develop propositions for a variety of case scenarios to address the probability of the evidence if the DNA originated or did no originate from one or more persons of interest.
6. Develop conditional propositions for use when the DNA of a known individual can be reasonably assumed to be present on the evidence.

Presenters

• Dr. David Balding | University of Melbourne, Melbourne, Australia
• Dr. Michael Coble | University of North Texas Health Science Center, Fort Worth, Texas
• Dr. John Buckleton | Institute of Environmental Science and Research, Auckland, New Zealand
• Steven Myers | California Department of Justice, Richmond, California

Recommended Reading

• Balding, D. and Steele, C. “Weight of Evidence for Forensic DNA Profiles”, Wiley, 2nd ed., 2015.
• Gittelson, S., Kalafut, T., Myers, S., Taylor, D., Hicks, T., Taroni, F., Evett, I.W., Bright, J.-A., and Buckleton, J., A Practical Guide for the Formulation of Propositions in the Bayesian Approach to DNA Evidence Interpretation in an Adversarial Environment, J. Forensic Sci., 2016, 61(1): 186-195.
• Ramos, D. and Gonzalez-Rodriguez, J., Reliable support: Measuring calibration of likelihood ratios, Forensic Sci. Int., 2013, 230, 156–169.
• Slooten, K. and Caliebe, A., Contributors are a nuisance (parameter) for DNA mixture evidence evaluation, Forensic Sci. Int. Genet., 2018, 37: 116-125.