hello all, I have a SNP genotype data of 228 accessions of cotton, on which I want to do GWAS. The genotype data has a lot of missing SNPs, which should be imputed. I have calculated genotyping rate in plink, which is 0.72. My goal is to do GWAS. The genotype data is already published, and I downloaded it from a link given in the supplementary portion of the paper. They imputed it from beagle, but when I tried to contact them, they didn't answer me. The format of genotypic data they published, does not follow any format, neither vcf nor hapmap or any other format. I converted it in hapmap format manually for association in GAPIT and TASSEL. But Beagle does not support hapmap, so I converted in VCF format with TASSEL, which also looks like wrong. To impute the missing data, I run imputation in beagle, and got the results. But as far as I read in the literature, a reference panel is used for imputation. But in the case some crops, where there is no any reference panel. What to do? I run the imputation without any reference panel, and I am not sure, whether my imputed results are right or wrong. Need suggestions.
STITCH is great, but requires the read sequencing data in BAM or CRAM, sounds like the OP has just the called genotypes?