Distinguishing between donors and their relatives in complex DNA mixtures with binary models

Area of Science:

• Forensic Science
• Genetics
• Statistical Analysis

Background:

• Traditional likelihood ratio calculations for DNA mixtures were restricted to profiles with complete allele data.
• Recent advancements enable analysis of DNA mixtures with allelic dropout and drop-in, expanding applicability but altering interpretation requirements.
• This necessitates re-evaluating the significance of partial matches, especially concerning relatives of potential contributors.

Purpose of the Study:

• To investigate the consequences of advanced DNA mixture analysis on likelihood ratios involving relatives of donors.
• To assess how assuming uniform dropout probabilities impacts false positive probabilities for relatives compared to donor-specific probabilities.

Main Methods:

• Utilized a semi-continuous binary approach for likelihood ratio calculations.
• Incorporated observed alleles and donor-specific dropout probabilities.
• Did not rely on peak height information, focusing on allele presence and dropout events.

Main Results:

• The study explores the impact of simplified dropout probability models on false positive rates.
• It quantifies how deviations from accurate dropout probabilities can affect the interpretation of evidence involving relatives.

Conclusions:

• Advanced DNA mixture analysis, while powerful, requires careful consideration of dropout probabilities.
• Inaccurate dropout probability assumptions can significantly alter the assessment of evidence, particularly for relatives of donors in forensic investigations.