Additional file 1 of Environmental differences explain subtle yet detectable genetic structure in a widespread pollinator

Additional file 1: Table S1. Microsatellite loci used in this work, with original locus names [127], the primer mixes they were assigned to, repeat units (motifs), size ranges, annealing temperatures (Tm), forward and reverse primer sequences (5’–3’) as well as information on null alleles and stuttering. Table S2. Basic population genetic statistics for each sampling location encompassing diploid (‘dpds’) and both haploid and diploid individuals (‘mpds’). N: number of individuals; NA: number of alleles; NE: number of effective alleles; AR: allelic richness, rarefied for seven individuals (k); uHE: unbiased expected heterozygosity; HO: observed heterozygosity. Observed heterozygosity values between ‘dpds’ and ‘mpds’ are largely congruent, as SPAGeDi, even though capable of working with mixed-ploidy data sets, does not include haploid individuals in calculating this metric. Table S3. Below the diagonal: Multilocus pairwise Fst values calculated with GenAlEx and based on a data set containing diploid individuals only. Above the diagonal: Corresponding FDR-corrected P values. Marginally significant Fst values and P values are indicated in bold. See Table S8 for population names. Table S4. Above the diagonal: Multilocus pairwise Fst values computed with polysat and based on a data set containing both haploid and diploid individuals. Below the diagonal: Corresponding bootstrap-derived upper 95% confidence intervals. See Table S8 for population names. Table S5. Above the diagonal: Multilocus pairwise Fst values calculated with polysat and based on a data set containing diploid individuals only. Below the diagonal: Corresponding bootstrap-derived upper 95% confidence intervals. See Table S8 for population names. Table S6. Discriminant analysis of principal components (DAPC) result showing the assignment of individuals to clusters identified de novo. Table S7. Pearson correlation coefficients of environmental variables at population locations harbouring both haploid and diploid individuals. See Table S9 for the abbreviations used. Table S8. Populations sampled, including their abbreviations, locations, and the number of individuals. ‘cpds’: full data set after exclusion of full siblings and clones; ‘pop10’: ‘cpds’ after excluding populations with fewer than 10 individuals; ‘mpds’: mixed-ploidy data set encompassing both haploid males and diploid females; ‘dpds’: data set containing diploid individuals only. Table S9. Environmental variables used. Those in bold were retained after stepwise elimination with a variance inflation factor ≥ 10. Table S10. Resistance matrix derived from Circuitscape, where higher values indicate higher predicted resistance to inter-population dispersal. See Table S8 for population names.