A NOVEL MEASURE OF GENETIC-DISTANCE FOR HIGHLY POLYMORPHIC TANDEM REPEAT LOCI

Genetic distance measures are indicators of relatedness among populati ons or species and are useful for reconstructing the historic and phyl ogenetic relationships among such groups. Classical measures of geneti c distance were developed to analyze biochemical and serological polym orphisms, systems which generally show limited variability. However, t hese traditional measures of genetic distance are inadequate for the a nalysis of certain classes of variable number tandem repeat (VNTR) loc i, which have a larger number of alleles and higher levels of heterozy gosity than traditional genetic markers. At the higher levels of heter ozygosity observed at these loci, the standard measures of genetic dis tance are nonlinear and do not account for the mutational mechanisms o f hypervariable loci. We have developed a measure of genetic distance, D-sw, which is appropriate for the analysis of highly polymorphic DNA loci. Using computer simulations of diverging populations, we show th at D-sw conforms to linearity and that the variance is similar in magn itude to traditional measures of genetic distance. Comparisons of phyl ogenetic trees derived from the simulated divergence of human racial g roups demonstrate that the branch lengths of trees prepared using D-sw are more similar to the model tree than those generated using other m easures. Finally, we demonstrate the applicability of D-sw to evolutio nary analysis by reconstructing the relationships among eight human po pulations using 14 microsatellite and STR loci. The phylogenetic trees generated using D-sw are different from trees constructed with tradit ional measures and better reflect the well-documented ancient divergen ce of African and non-African populations.