Can simple population genetic models reconcile partial match frequencies observed in large forensic databases?

Area of Science:

• Forensic Science
• Population Genetics
• Statistical Modeling

Background:

• Forensic DNA databases are crucial for criminal investigations.
• Partial matches in DNA profiles can occur due to database size and genetic similarities.
• Understanding match frequencies is vital for accurate interpretation of forensic evidence.

Purpose of the Study:

• To model the expected number of partial DNA profile matches in large databases.
• To investigate the implications of population substructure and relatedness on match frequencies.
• To evaluate the plausibility of observed partial matches in the Arizona offender database.

Main Methods:

• Utilized simple statistical models incorporating the product rule.
• Incorporated parameters for population substructure and genetic relatedness.
• Applied models to predict expected matches within a large DNA offender database.

Main Results:

• A significant number of nine and ten locus partial matches were identified in the Arizona database.
• The study found a narrow range of parameter values that could plausibly explain the observed Arizona results.
• Model predictions suggest specific conditions under which such matches are expected.

Conclusions:

• The observed partial DNA matches in Arizona may be explained by current statistical models within specific parameter constraints.
• Further research is needed to validate these models against empirical data.
• Adjustments to fundamental assumptions in predicting DNA match frequencies may be necessary.