If two populations are involved, the possibilities are nucleotide versus nucleotide, nucleotide versus haplotype, and haplotype versus nucleotide. Each possibility can be tested after sampling using t-test. If more than two populations are involved then analysis of variance (ANOVA) may be involved.

It depends on the number of population groups. You can use student t-test if you have only two groups otherwise, you do a post-hoc analysis. There are quite a number of tests like the LSD, Turkey, Scheffe etc. It depends on your goal and also whether you have paired or independent grouping of population.