I recently ran a GWAS in GEMMA to fit a linear mixed model and generate p-values using the likelihood ratio test. I've identified 20 SNPs that are statistically significant. The output file just includes the following columns: SNP ID, chromosomal position, number missing in my dataset, major allele, minor allele, minor allele frequency, "logl_H1," "l_mle," and finally the p-values. Is there a way that I can figure out the odds ratios for the significant SNPs given this information? My genome files are in PLINK binary format, so I'm thinking there must be some way to use PLINK to calculate odds ratios by inputting the significant SNP ID's, but I haven't come a cross a way to do this.
Edit: I tried running a second GWAS in PLINK and just using the odds ratios that generated. However this only reports the raw ORs and I need to be able to adjust for some covariates. For that I'd need some way to know which individuals have the associated SNP variant so that I can run a logistic regression. Any suggestions?
Is there no column in your GEMMA output called 'beta'? - this would relate to the beta coefficient, the exponent of which is the odds ratio.