Name & behavior & side-effect handling of `null_weight`

stephenslab / susieR

R package for "sum of single effects" regression.

https://stephenslab.github.io/susieR

Other

179 stars 47 forks source link

Name & behavior & side-effect handling of `null_weight` #36

Closed gaow closed 6 years ago

gaow commented 6 years ago

Just open this ticket as a self-reminder that we need to re-visit the (experimental) null_weight parameter after I've updated my DSC to have it evaluated.

gaow commented 6 years ago

And that the summary statistics version needs be updated too when it comes to that point. @zouyuxin you can then compare a future version with version 0.4.1 to see what has been changed in between for the full version.

gaow commented 6 years ago

Currently susie_get_CS and the new function summary() both removes null by default (a behavior which cannot be configured but can be hacked). Yuxin and I have also agreed to reuse my init_setup, where the susie object gets marked the null variable coordinate. So this problem should be taken care of automatically in the summary statistics version.