ISHI On-Demand: Probabilistic Genotyping, DNA Mixtures, and Likelihood Ratios
Available for an introductory price of $25!
Probabilistic genotyping has become a central tool for interpreting complex DNA mixtures, yet its adoption continues to raise scientific, operational, and communication challenges for forensic laboratories. Questions around validation design, number-of-contributors assumptions, robustness, reporting practices, and courtroom explanation remain active areas of discussion across jurisdictions.
This ISHI On-Demand module brings together leading forensic scientists, statisticians, and implementation experts to share practical perspectives drawn from casework experience, research, and laboratory oversight. Through a series of moderated conversations, the speakers explore how probabilistic genotyping fits into modern forensic workflows, where its strengths and limitations lie, and how analysts and laboratory leaders can approach interpretation and communication with transparency and scientific rigor.
Featured speakers include:
- Richard Wivell, Senior Scientist, New Zealand Institute for Public Health and Forensic Science (PHF Science)
- Jo Bright, Senior Science Leader, New Zealand Institute for Public Health and Forensic Science (PHF Science)
- Dr. Michael Coble, Executive Director, Center for Human Identification, University of North Texas
- Austin Hicklin, Director, Forensic Science Group, NOBLIS
- Nicole Richetelli, Forensic Statistical Analyst, NOBLIS
- Cydne Holt, Principal & Chief Scientific Officer, Holt BioForensic Experts
Total video time: 51 minutes, 37 seconds
Module-Level Learning Objectives
After engaging with this ISHI On-Demand module, participants will be able to:
- Describe how probabilistic genotyping is used to support interpretation of complex DNA mixtures, including its relationship to analyst judgment.
- Identify common scientific and operational challenges laboratories encounter when implementing and maintaining probabilistic genotyping workflows.
- Explain how assumptions such as number of contributors, conditioning, and model parameters influence interpretation outcomes.
- Recognize the role of validation, revalidation, and documentation in supporting transparency and defensibility.
- Articulate key considerations when communicating likelihood ratios and probabilistic results to non-scientific audiences, including courts.
- Reflect on emerging developments in mixture interpretation and probabilistic methods and how they may affect future laboratory practice.
Video 1: Foundations of Probabilistic Genotyping and DNA Mixtures
Speakers:
Richard Wivell, Jo Bright, Dr. Michael Coble
Description:
This opening segment establishes a shared foundation for probabilistic genotyping by examining why DNA mixtures remain a persistent challenge for forensic laboratories and how probabilistic approaches are used to address those challenges. The speakers discuss probabilistic genotyping as a tool that supports, rather than replaces, analyst expertise, highlighting how likelihood ratios, model assumptions, and data trends are integrated into interpretation. Limitations, including complex mixture scenarios and closely related contributors, are also addressed to reinforce the importance of scientific judgment and context.
Video 2: Validation and Implementation Considerations for Probabilistic Genotyping
Speakers:
Richard Wivell, Jo Bright, Dr. Michael Coble
Description:
Focusing on validation and implementation, this segment explores how laboratories approach probabilistic genotyping adoption in practice. Topics include defining validation plans, learning from laboratories that have implemented earlier, and balancing rigor with efficiency. The discussion also addresses how validation studies relate to real casework, how laboratories evaluate software updates, and how validation results can be communicated transparently when ground truth is unknown.
Video 3: Interpretation, Assumptions, and Robustness in Probabilistic Genotyping
Speakers:
Richard Wivell, Jo Bright, Dr. Michael Coble
Description:
This segment examines how interpretive assumptions influence probabilistic genotyping results. The speakers discuss number-of-contributors estimation, laboratory-specific interpretation limits, conditioning on known contributors, and the risks of overconfidence when ground truth cannot be established. Emphasis is placed on understanding how assumptions affect likelihood ratios and on maintaining transparency when evaluating competing explanations for complex mixtures.
Video 4: Communicating Likelihood Ratios and Probabilistic Results in Court
Speakers:
Richard Wivell, Jo Bright, Dr. Michael Coble
Description:
Effective communication of probabilistic results is a recurring challenge in forensic science. In this segment, the speakers share perspectives on explaining likelihood ratios to non-scientific audiences, including judges and juries. Analogies, transparency around assumptions, and discussion of uncertainty are explored alongside considerations for cross-examination. The segment also introduces activity-level evaluations, highlighting differences in adoption between jurisdictions and the role of training and guidance.
Video 5: Lessons from the Field: Implementing and Maintaining Probabilistic Genotyping
Speakers:
Austin Hicklin, Nicole Richetelli
Description:
Drawing from a national effort to collect lessons learned, this segment focuses on the practical realities of probabilistic genotyping implementation. The speakers discuss common challenges laboratories face, including resourcing, timelines, training, documentation, and revalidation. Rather than prescribing solutions, the conversation highlights recurring themes observed across laboratories and emphasizes the importance of realistic planning, adaptability, and shared community experience.
Video 6: Mixture Interpretation Beyond Traditional Workflows: Perspectives from Research and Casework
Speaker:
Cydne Holt
Description:
This segment explores mixture interpretation through the lens of emerging data types and analytical approaches. Drawing on research and applied casework, the discussion examines limitations of traditional methods, the value of sequence-based information, and how automated or semi-automated tools can support documentation, review, and transparency. The segment provides perspective on how evolving technologies may influence mixture interpretation while reinforcing the importance of clear logic and defensible reasoning.
Video 7: Looking Ahead: The Future of Probabilistic Genotyping and Analyst Practice
Speakers:
All
Description:
In this closing segment, the speakers reflect on how probabilistic genotyping and mixture interpretation may continue to evolve. Topics include emerging statistical approaches, automation and AI, analyst training, and long-term implementation considerations. The discussion reinforces the need for continued learning, critical thinking, and adaptability as forensic science advances.