Analysis of population genetic data often includes the search for genomic regions with signs of recent positive selection. One of the approaches involves the concept of Extended Haplotype Homozygosity (EHH) and its associated statistics. These statistics typically need phased haplotypes and, some of them, polarized variants. Here, we unify and extend previously proposed modifications to loosen these requirements. We compare the modified versions with the original ones by measuring the False Discovery Rate on simulated whole-genome scans and quantifying the overlap of inferred candidate regions on empirical data. We find that phasing information is indispensable for the accurate estimation of within-population statistics for all but very large samples and of cross-population statistics for small samples. Ancestry information, in contrast, is of lesser importance for both. Our publicly available R package rehh incorporates the modified statistics presented here.

Analysis of population genetic data often includes the search for genomic re- gions with signs of recent positive selection. One of the approaches involves the concept of Extended Haplotype Homozygosity and its associated statistics. These statistics presume that haplotypes are phased and some of them that variants are polarized. Here we assess the consequences if one of the two con- ditions is not fulfilled. We find that phasing information is indispensable for the accurate estimation of within-population statistics and, if sample sizes are small, for cross-population statistics, too. Ancestry information, in contrast, is of lesser importance for both. We make use of a publicly available update of our R package rehh which, among other features, incorporates the adapted statistics presented here.