Hi,
Is the "est" (estimated effect) always a standardised beta coefficient in userGWAS() output?
For context in case it's needed:
- I'm doing a GWAS-by-subtraction with two traits
- Both original traits sumstats are case/control studies
- Trait1 GWAS used logistic model, trait2 had binary outcome but z-statistics in sumstats only
- The sumstats () output I created, which is then used in userGWAS, has beta.trait1 and beta.trait2
Many thanks,
Aoife