How to determine which phenotype the SNPS found by MOST belong to?

[denvdm]

Hi, I suggest you simply take a look at the univariate GWAS results

[ThisHq]

Hi, I suggest you simply take a look at the univariate GWAS results

thanks，I also want to ask what to do with covariate files, I don't see covariate files as input for Mostest

[denvdm]

The outcome measures should be pre-residualized, please see the 'other considerations' section in the readme.

[ThisHq]

The outcome measures should be pre-residualized, please see the 'other considerations' section in the readme.

Thanks，I saw the notes on pre-residualized, and I would like to ask whether the code for pre-residualized is resid.m? I saw in the notes that R is used to accomplish this, and I don't know which method can accomplish pre-residualized. Thank you for your answer.

[denvdm]

No method included, this is indeed done in R with straightforward linear regression, as exemplified in the 'other considerations' section. In other words, you'll need to adapt this code yourself to your specific data/covariates, before running MOSTest:

for(f in outcomes){ tmplm <- summary(lm(get(f) ~ Age + Sex + Scanner + OtherCovars, data=df)) df[,which(colnames(df)==f)] <- tmplm$residuals }

hope this helps

[ThisHq]

No method included, this is indeed done in R with straightforward linear regression, as exemplified in the 'other considerations' section. In other words, you'll need to adapt this code yourself to your specific data/covariates, before running MOSTest:

for(f in outcomes){ tmplm <- summary(lm(get(f) ~ Age + Sex + Scanner + OtherCovars, data=df)) df[,which(colnames(df)==f)] <- tmplm$residuals }

hope this helps Thanks，I tried to implement it using R, which generates a list of residuals. I don't know if I got this right, “f” is the phenotype and “df” is the covariate. But I don't quite understand "df[,which(colnames(df)==f)]". "which (colnames(df)==f)" stands for finding the same columns in phenotypes and covariates, what do these columns refer to? FID, IID? Is it correct that I replace the original phenotype with the generated list of residuals and input it into Mostest as the new phenotype？ Sorry to bother you again.