Suzanne Jin

Hello @grst , currently we have done: - [x] #383 - [x] #423 The next step would be: - [ ] #422

> Thanks @suzannejin - another big one. Will review thoroughly when I can, just give me some time :-) Unfortunately many parts of the pipeline had to be changed to...

> Just tackled the utils subworkflow so far, we need to find some way to make the logic easier to follow. I've made a suggestion for one possibility, using functions...

> With the toolsheet implemented, can we reduce the number of input parameters because they are redundant now? E.g., do we still need > > * `study_type` > > *...

> I've been thinking deeply about the current implementation and have a proposal to significantly simplify our approach while maintaining functionality. > > > > ### Core Proposal: Simplify the...

Hello @pinin4fjords, as for the discussion we had about deduplication and avoiding modules to run again unnecessarily, it seems to me that this will not only happen for a differential...

> [@suzannejin](https://cbcrg.slack.com/team/UV9NYQVK6) I think we will need to collapse-and-expand wherever we want to explicitly avoid duplicated computation and then restore the link of a result to a particular analysis. >...

> Hello! I had a first look at the suggestion of providing all parameters through the toolsheet. While I am not completely against it, I am a bit worried that...

> > We could do the collapse-and-expand on every module, but it does add a bit of noise to the code, so it makes sense to hold off until we...

> We could do the collapse-and-expand on every module, but it does add a bit of noise to the code, so it makes sense to hold off until we actually...