The filtered SNP dataset (VCF file)

We collected 66 individuals for genome resequencing from 18 populations across its range. Each sample produced approximately 10× coverage for the reference genome.

Trimmomatic v0.39 was used to control the quality of raw reads with default parameters. Clean reads were aligned using the BWA-MEM algorithm of GATK v.4.2 to improve the alignment around putative indels. We performed a series of critical steps for filtering single-nucleotide polymorphisms (SNPs) to reduce false positives. After stringent site- and genotype-level filtering, we retained 39,178,374 high-confidence biallelic SNPs (The filtered SNP dataset) for downstream analyses.