Genetic structure of the Utah Mormons: comparison of results based on RFLPs, blood groups, migration matrices, isonymy, and pedigrees.

The genetic structure of the Utah Mormon population is examined using 25 blood group and 47 RFLP alleles obtained from 442 subjects living in 8 geographic subdivisions. Nei's GST was 0.013 (p < 0.002) for the RFLP data and 0.012 (p > 0.4) for the blood group data, showing that only 1% of the genetic variance in this population can be attributed to subdivision effects. A comparison of intersubdivision distance matrices based on blood groups, RFLPs, migration matrices, isonymy, and pedigrees shows that genetic distances have relatively low and nonsignificant correlations with the other three types of data. However, the correlations based on RFLPs are considerably higher than those based on blood groups. Relationship matrices based on interindividual allele sharing were compared with known genealogical kinship coefficients between each pair of individuals. The correlation between the blood group and RFLP relationship matrices was small but marginally significant using the Mantel test (r = 0.014, p < 0.06). The RFLP relationship matrix correlated more highly with genealogical kinship than did the blood group relationship matrix (r = 0.023, p < 0.0001 and r = 0.012, p < 0.001, respectively). These correlations increased by approximately one order of magnitude when pairs of subjects having zero kinship coefficients were excluded. These results show that genetic distances derived from RFLPs correlate more strongly with other types of kinship than do distances based on blood groups. This probably reflects the fact that RFLPs are more neutral, have frequencies that are more accurately estimated, and contain more information about DNA sequence variation.