A method for interpreting mixed DNA evidence based on the gamma model

Area of Science:

• Forensic Science
• Genetics
• Computational Biology

Background:

• Mixed DNA evidence from crime scenes is crucial but challenging to interpret.
• Current methods struggle to precisely identify multiple contributors and their proportions in complex DNA mixtures.
• Advanced forensic genetic technologies still face interpretation bottlenecks.

Purpose of the Study:

• To develop a novel algorithm for accurate interpretation of complex multi-contributor DNA profiles.
• To address the limitations of traditional methods in analyzing mixed DNA evidence.
• To enhance the reliability and practical utility of forensic DNA analysis.

Main Methods:

• A continuous gamma distribution model based on probabilistic residual optimization was developed.
• A two-step probabilistic framework was employed, involving allelic permutations and preliminary proportion estimation.
• Iterative maximum likelihood estimation and gamma distribution were used to optimize genotype combinations and contributor proportions simultaneously.

Main Results:

• The algorithm successfully generates candidate genotype combinations and estimates contributor proportions.
• It dynamically optimizes parameters (α and β) to calculate residual probability weights.
• Maximum likelihood solutions are derived by integrating population allele frequency databases, providing accurate results.

Conclusions:

• The proposed algorithm offers a reliable and quantifiable tool for forensic identification.
• It significantly improves the accuracy of complex mixed DNA profile interpretation.
• This advancement enhances the practical utility of mixed DNA in criminal investigations and supports judicial fairness.