Likelihood Ratios Given Activity-Level Propositions for DNA Transfer Evidence: Practical Implementation and...

Area of Science:

• Forensic Science
• Genetics
• Statistical Modeling

Background:

• Interpreting low-template DNA requires models accounting for lab variability, transfer uncertainty, and case specifics.
• Activity-level likelihood ratios (LRs) are crucial for DNA evidence interpretation at the activity level.

Purpose of the Study:

• To present the practical implementation of HaloGen, an open-source hierarchical Bayesian framework for calculating activity-level LRs from DNA quantity data.
• To compare three modeling approaches (Group, Lab-Bayes, Lab-Vague) within the HaloGen framework.
• To propose a calibration pathway for laboratories using HaloGen for quantitative LR reporting.

Main Methods:

• Developed and implemented HaloGen, a hierarchical Bayesian framework.
• Compared three modeling approaches: Group, Lab-Bayes, and Lab-Vague.
• Utilized simulation studies to assess evidential strength sensitivity to DNA quantity and case context.

Main Results:

• Evidential strength depends on DNA quantity, case context (number of actors, specified unknowns), and laboratory calibration.
• Inter-laboratory differences in DNA recovery and non-detection can significantly alter LRs.
• Limited local transfer experiments can improve relevance over pooled population models.

Conclusions:

• HaloGen provides a framework for quantitative activity-level LR calculation, accommodating inter-laboratory variability.
• Transparent specification of propositions, data relevance, model assumptions, and expert judgment is critical for activity-level inference.
• A minimum-effort calibration pathway is proposed for laboratories using HaloGen.