Incorporating Relatedness Information Into the Interpretation of DNA Mixtures in Complex Forensic Investigation Scenarios Devoid of Persons of Interest

Most current forensic applications concerning DNA mixtures primarily focus on individual identification, specifically determining the presence of a person of interest (POI). However, when the target individual is unavailable due to death or disappearance, traditional methods often prove inadequate. Our previous exploratory research has demonstrated that integrating Bayesian network algorithms with the novel genetic marker-microhaplotype-typing technology can effectively facilitate forensic DNA mixture analysis across various scenarios. However, it is important to note that short tandem repeat (STR) typing based on capillary electrophoresis (CE) remains the most widely employed method in forensic practice, and the profiling data for the majority of DNA mixed samples are still derived from this established technology. Therefore, further investigation into various scenarios devoid of POIs based on traditional STR genotyping data is warranted. In this study, we undertook an investigation into three scenarios involving DNA mixtures in the absence of POIs, leveraging relatedness information through Bayesian network algorithms. The analyses were based on traditional CE-based STR genotyping data derived from artificially synthesized and simulated mixed samples. The Bayesian network framework offers considerable flexibility, enabling the incorporation of kinship information for various interpretive purposes, including assessing the potential contribution of a missing pedigree member to a DNA mixture and evaluating the relatedness among contributors within or between mixed DNA profiles. The aforementioned research offers a referable experimental framework for addressing complex DNA mixtures within the realm of forensic practice.