2.4 | Population genetic analyses
We selected high-quality SNPs which contained no more than 20% missing data, and thinned the sites such that no two sites were within the same 2000-bp region. Thereby, we eliminated the potential effects of physical linkage among variants. We then conducted a phylogenetic analysis with the final SNP set using the IQ-TREE software (version 1.6.9). We constructed a maximum likelihood (ML)-based phylogenetic tree using the GTR + F + R5 model, and ran 1000 rapid bootstrap replicates to determine the confidence values of phylogenetic branches.
To visualize patterns of genetic variation, we conducted a principal component analysis (PCA) of the final SNP set using PLINK (version 1.90), and plotted the resultant principal components against one another using the software R (version 3.4). we also used the final SNP set for a population structure analysis using ADMIXTURE (version 1.3), which was ran with K values (the number of assumed ancestral components) ranging from 1 to 10.