DNA extraction and sequencing
From the tissue samples, we purified DNA for library preparation using the Biotools Speedtools Tissue DNA Extraction kit. Next, we used the restriction enzyme Pst1 to prepare the libraries for genome-wide genotyping with GBS (Genotyping by Sequencing ). The sequencing was done in an Illumina HiSeq2500 sequencer, and once the sequences were obtained we characterized single nucleotide polymorphisms (SNPs). For this purpose, we used the UNEAK pipeline, implemented in TASSEL v.3.0 (Bradbury et al. 2007), and specifically designed for samples of species without a reference genome. The sequences were pruned to eliminate sequencing errors using the error rate threshold parameter. The resulting database consisted of 83,648 SNPs with a coverage of 5.68 ± 6.56 (mean ± sd) and missing site rate of 0.49 ± 0.33. We then discarded loci with minor allele frequencies < 0.01, and those that were sequenced in less than 10% of individuals. The final database consisted of 73,291 loci, with a single SNP per locus and a coverage of 6.6 ± 6.75 and a missingness of 0.42 ± 0.31. This dataset was the one used to analyze the neutral variation of our samples.