I would like to use GWAS to detect genetic differences between subpopulations of bacteria, using subpopulations as the phenotype. In this case, is it necessary to correct for population structure?

Actually, I tried GWAS using pyseer to detect lineage specific unitigs, but it seemed that these included a lot of false positive unitigs, according to qqplot.

Before analysis, I performed population structure analysis, and definded 2 populations. Then, I used these population as trait(pop1 ->0 pop2 ->1 in trait file).

Best regards,