Minimal Y-chromosomal haplotype selection for phylogenetic study using the bootstrapped DTI method

Abstract

In our previous report, a decision tree induction (DTI) algorithm proved to be an effective tool to measure impurities of categorizing populations and to select the minimum number of Y-microsatellite markers, with the highest discriminating power. By using the DTI algorithm to determine the most informative loci in four studied hilltribe populations, i.e. the Akha, Hmong, Karen, and Lisu, seven powerful loci for phylogenetic reconstruction were selected and validated. However, the accuracy of the DTI algorithm on haplotype selection of other hilltribe populations was still in question. Therefore, a bootstrap method was applied in this study to the same data and contexts to verify the results of the DTI algorithm. We also examined the bootstrap method in a larger sample size and with a larger number of populations. Three informative microsatellite loci were selected to differentiate among the populations. When all loci were ranked, the three selected loci usually appeared at high potential levels. It can be concluded that we were able to regenerate the minimal haplotype- a combination of three selected loci- with a high level of confidence for effective phylogenetic study of human populations.