Differences in urbanization degree and consequences on the diversity of conventional vs. rapidly mutating Y-STRs in five municipalities from a small region of the Tyrolean Alps in Austria.

In this study we set out to test at a micro-geographic scale for the potential effects of differences in urbanization degree on Y-chromosomal diversity and the paternal lineage differentiation of "conventional" and rapidly-mutating (RM) Y-STR markers. To avoid systematic underrepresentation of common lineages, 551 male samples were collected under a sampling regime allowing for the inclusion of paternal relatives. All participants came from a small, topographically highly structured, yet culturally homogeneous settlement area in the Tyrolean Alps of Austria, a region that is characterized by a longstanding coexistence of communities differing considerably in size and connection. The study participants reported provenance in one of the three rural villages Alpbach, Brandenberg, and Wildschönau - all being separated by topographical barriers from each other - or in one of the two more urban-like and better connected municipalities Kitzbühel and St. Johann in Tirol. When compared with the sample pools from the two larger communities, the three small villages showed distinctly higher rates of self-reported patrilocality since the paternal grandfather (85-95% vs. ∼42%), and featured evidence for a considerably higher proportion of close and cryptic paternal relationships among the study participants. We observed marked differences in the Y-SNP haplogroup frequency spectra and statistically significant Y-STR-based FST distances among the municipality samples, suggesting population sub-structuring along municipality borders. While for the two larger settlements a widely used "core" set of 17 conventional Y-STRs (Yfiler) provided reasonably high lineage resolution (Ĥ: 0.99515±0.00256, 0.99739±0.00224), a markedly reduced haplotype diversity was seen in samples from the rural villages (Ĥ: 0.96126±0.00701-0.98515±0.00278). This difference largely diminished when instead using a set of 13 RM Y-STRs (Ĥ: 0.99180±0.00380-0.99922±0.00187, for all groups). Most notably, in the Alpbach sample the number of different haplotypes rose from 42 (Yfiler) to 99 (RM Y-STRs) and the proportion of matching haplotypes dropped from nearly 4% (Yfiler) to about 0.4% (RM Y-STRs) of the pairwise comparisons. Consistent results were obtained with a reduced version of the dataset, being devoid of inferred close male relatives up to the degree of first cousins. Finally, consequences potentially arising from a gain in lineage-resolution for population reference-sample size requirements will be addressed briefly.