Using Bayesian networks to track DNA movement through complex transfer scenarios.

Area of Science:

• Forensic Science
• Molecular Biology
• Computational Biology

Background:

• Trace DNA transfer is crucial in forensic investigations and legal proceedings.
• Evaluating the probability of DNA findings requires understanding transfer mechanisms.
• Previous work established a basic Object-Oriented Bayesian Network (OOBN) for trace DNA analysis.

Purpose of the Study:

• To expand the existing OOBN for trace DNA analysis.
• To formalize class networks for modeling complex DNA transfer chains.
• To enable logical consideration of DNA movement through sequential transfer events.

Main Methods:

• Development of an expanded Object-Oriented Bayesian Network (OOBN).
• Formalization of class networks to model two-way DNA transfer between items.
• Inclusion of sampling for intermediary items in transfer chains.

Main Results:

• The enhanced OOBN logically integrates elementary building blocks for complex transfer scenarios.
• The model accommodates two-way DNA transfer upon item contact.
• Demonstrated application of the OOBN to a case scenario involving a chain of transfers.

Conclusions:

• The developed OOBN provides a flexible framework for analyzing complex trace DNA transfer events.
• This approach allows for tackling diverse forensic cases by combining basic network components.
• The study validates the OOBN's utility in real-world case scenarios.