A methodology for unsupervised clustering using iterative pruning to capture fine-scale structure
SNP-based information is used in several existing clustering methods to detect shared genetic ancestry or to identify population substructure. Here, we present a methodology for unsupervised clustering using iterative pruning to capture fine-scale structure called IPCAPS. Our method supports ordinal data which can be applied directly to SNP data to identify fine-scale population structure. We compare our method to existing tools for detecting fine-scale structure via simulations. The simulated data do not take into account haplotype information, therefore all markers are independent. Although haplotypes may be more informative than SNPs, especially in fine-scale detection analyses, the haplotype inference process often remains too computationally intensive. Therefore, our strategy has been to restrict attention to SNPs and to investigate the scale of the structure we are able to detect with them. We show that the experimental results in simulated data can be highly accurate and an improvement to existing tools. We are convinced that our method has a potential to detect fine-scale structure.