A total of 79 datasets (including 62 previously reported wild soybean genomic datasets, https://www.ncbi.nlm.nih.gov/sra/?term=(SRP045129)+AND+%22Glycine+soja%22%5Borgn%3A__txid3848%5D+++%EF%BC%89) were utilized for further analysis. Sentieon was used to detect SNPs and InDels, which were subsequently annotated using ANNOVAR software. SNPs with a RMS Mapping Quality (MQ) below 40, genotype quality (GQ) below 5 and a genotype depth (DP) below 4 were redefined as missing. After SNP filtering with the following conditions: QC<20, MAF<0.05, missing>0.2, 9,686,569 SNPs were obtained. Indels were similarly analyzed under the same criteria, resulted in a total of 1,650,281 indels.