Strategies for pairwise searches in forensic kinship analysis.

Area of Science:

• Forensic Genetics
• Population Genetics
• Bioinformatics

Background:

• Kinship testing is crucial for diverse applications, from forensic investigations to archaeological studies.
• Existing methods often have limitations in handling complex relationships, inbreeding, and non-autosomal markers.
• Large-scale DNA testing increases the risk of errors, necessitating robust statistical approaches.

Purpose of the Study:

• To develop generalized methods for pairwise kinship testing applicable to general and inbred populations.
• To incorporate non-autosomal markers, such as X-chromosomal markers, for enhanced relationship resolution.
• To address the challenges of multiple testing and dependent likelihood ratios in large-scale kinship analyses.

Main Methods:

• Modeling general, possibly inbred, pairwise relationships.
• Utilizing autosomal and X-chromosomal markers.
• Applying the theory of multiple testing to determine optimal thresholds.
• Incorporating a Bayesian approach for robust inference.
• Developing methods based on freely available software.

Main Results:

• The generalized approach successfully models complex kinship scenarios, including inbred populations.
• The inclusion of X-chromosomal markers allows for distinguishing specific relationships like half-siblings.
• Methods for determining optimal thresholds effectively manage the increased error risk in multiple testing.
• The Bayesian approach provides a valuable framework for interpretation.

Conclusions:

• The developed methods offer a significant advancement in DNA kinship testing, particularly for complex and large-scale applications.
• The approach is versatile, applicable to forensic science, archaeology, and non-human population studies.
• The implementation, based on accessible software, ensures practical utility and broad adoption.