Linkage disequilibrium analysis of D12S391 and vWA in U.S. population and paternity samples

Recently, the European Network of Forensic Science Institutes voted to adopt five additional STR loci (D12S391, D1S1656, D2S441, D10S1248, and D22S1045) to their existing European Standard Set of seven STRs (TH01, vWA, FGA, D8S1179, D18S51, D21S11, and D3S1358). The D12S391 and vWA loci are located 6.3megabases (Mb) apart on chromosome 12. Ideally for use in forensic analyses, genetic markers on the same chromosome should be more than 50Mb in physical distance in order to ensure full recombination and thus independent inheritance. The purpose of this study was to evaluate if the closely located D12S391 and vWA loci are independent and, consequently, if these loci can be included in the product rule calculation for forensic and kinship analyses. Departures from Hardy-Weinberg equilibrium and linkage disequilibrium between the D12S391 and vWA loci were tested using n=654 unrelated U.S. African American, Caucasian, and Hispanic samples, and n=764 father/son paternity samples. In the unrelated U.S. population samples, no significant departures from HWE were detected for D12S391 or vWA. No significant evidence of linkage disequilibrium was observed between the loci in the population samples. However, significant linkage disequilibrium was detected in U.S. African American, Caucasian, and Asian father/son samples with phased genotypes. No significant linkage disequilibrium was detected for U.S. Hispanic paternity samples. The use of phased father/son pairs allowed for robust detection of linkage disequilibrium between D12S391 and vWA. In unrelated population samples, linkage disequilibrium is present but more difficult to detect due to the large number of possible haplotype combinations and unknown allelic phase. For casework analyses that involve unrelated or related individuals, the single-locus genotype probabilities for D12S391 and vWA should not be multiplied to determine the match probability of an autosomal STR profile. Since the D12S391 and vWA loci are not independent, it is recommended that the observed combination of alleles at D12S391 and vWA should be treated as a non-independent diplotype for profile probability calculations. The observed haplotype frequencies for U.S. African American, Caucasian, Hispanic, and Asian populations are provided for match probability calculations.