Microsatellite analyses
Allele dropout and stuttering measures were not significant and the presence of null alleles was indicated for all loci. Genotyping error was estimated to be 1.55%. Most pairwise comparisons indicated significant deviation from Hardy-Weinberg equilibrium (Supporting Information 3) and linkage disequilibrium (Supporting Information 4) after FDR corrections.
The DAPC (Fig. 1) presented clear structure composed by four populations (BRA+BP2+COL+ARG+URU; SIN+CUB; CHI+KOR+BEL; JAP). There is almost complete overlap among all South American countries, hereafter referred as Population 1 (Pop1), including the two populations described in Brazil (Jorgewich-Cohen et al., in press). Population 1 has a marginal link with Cuba, which is almost completely overlapped with samples from Singapore, together forming Population 2 (Pop2). Samples from the Republic of Korea, Belgium and China are also greatly overlapped and were considered as one unit (Pop3). Japan is completely isolated and was considered as a separated population (Pop4). Calculation of pairwise Jost’s D index values (Table 1) revealed greater differentiation between Pop3 and all others. The smallest differentiation reveled is between Pop2 and Pop4, followed by the differentiation between Pop1 and Pop2. The differentiation between pairs of populations was always significant (p=0.001). With exception for Pop2, the observed heterozygosity (Ho) did not match expected heterozygosity (He) and both populations showed positive inbreeding values (Supporting Information 3).