Y-chromosomal STR (Y-STR) profiles play a key role in forensic genetics, particularly in cases involving a DNA mixture with a major female contribution as commonly observed in sexual assaults. However, while the exclusion of non-matching suspects is straightforward, quantification of the evidential value of a Y-STR haplotype match has been a long-standing challenge in forensics. Up to now, the required Y-STR match probabilities have mostly been equated with the frequency of the relevant Y-STR profile in a population database even though defining a suitable ‘population’ for this purpose is non-trivial. 

We present a novel pedigree-based framework for calculating Y-STR match probabilities that splits the population of alternative donors into two, namely (i) the suspect’s close patrilineal relatives (his “pedigree”) and (ii) all other males. Approximation of both corresponding match probabilities proceeds by reconstructing and evaluating the Y-STR profiles of untyped pedigree members through simulation. The method has been implemented in a publicly available software tool, MatchY, that has since undergone extensive testing and validation. 

By explicitly accounting for the possibility that the trace donor is a close patrilineal relative of the suspect, our approach allows for a more realistic interpretation of the evidential value of Y-STR matches. In addition to describing its mathematical background and software implementation, we will demonstrate an application of the approach in scenarios relevant to practical forensic casework. 

Authors

• Amke Caliebe (Health and Medical University, Germany) 
• Manfred Kayser (Erasmus MC University Medical Center, Netherlands) 
• Michael Krawczak (Christian-Albrechts-University, Germany) 
• Arwin Ralf (Erasmus MC University Medical Center, Netherlands) 
• Dion Zandstra (Erasmus MC University Medical Center, Netherlands)