You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The runSingleTraitGwas() function ignores covariates present in the gData object if the "covar" argument is NULL (default).
This is counterintuitive and at this point I had to correct three colleagues that thought the covariates were being used.
The best interface choice imho should be to use by default all the covariates present in the gData. If you want to keep the current behavior at least put a warning (or a message) specifying that covariates are present but not used, and to use the "covar" argument to select what to include in the model.
The text was updated successfully, but these errors were encountered:
Thanks for your suggestion. I will discuss this with our internal users to see what their experience is with the "covar" argument and if a change in default behavior is needed.
A warning doesn't harm anyone, so that will definitely be added in the next version of the package.
The
runSingleTraitGwas()
function ignores covariates present in the gData object if the "covar" argument is NULL (default).This is counterintuitive and at this point I had to correct three colleagues that thought the covariates were being used.
The best interface choice imho should be to use by default all the covariates present in the gData. If you want to keep the current behavior at least put a warning (or a message) specifying that covariates are present but not used, and to use the "covar" argument to select what to include in the model.
The text was updated successfully, but these errors were encountered: