GENETIC DISTANCES AND RECONSTRUCTION OF PHYLOGENETIC TREES FROM MICROSATELLITE DNA

Recently many investigators have used microsatellite DNA loci for stud ying the evolutionary relationships of closely related populations or species, and some authors proposed new genetic distance measures for t his purpose. However, the efficiencies of these distance measures in o btaining the correct tree topology remains unclear. We therefore inves tigated the probability of obtaining the correct topology (P-c) for th ese new distances as well as traditional distance measures by using co mputer simulation. We used both the infinite-allele model (IAM) and th e stepwise mutation model (SMM), which seem to be appropriate for clas sical markers and microsatellite loci, respectively. The results show that in both the IAM and SMM CAVALLI-SFORZA and EDWARDS' chord distanc e (D-c) and NEI et al.'s D-A distance generally show higher P-c values than other distance measures, whether the bottleneck effect exists or not. For estimating evolutionary times, however, NEI's standard dista nce and GOLDSTEIN et at's (delta mu)(2) are more appropriate than othe r distances. Microsatellite DNA seems to be very useful for clarifying the evolutionary relationships of closely related populations.