search

NelleV / moanin

Timecourse transcriptomic analysis
https://nellev.github.io/moanin/
Other
5 stars 0 forks source link

Need to support bioconductor class #49

Closed epurdom closed 4 years ago

epurdom commented 4 years ago

I've opened this discussion to talk about how to add support for taking as input for the data argument the standard bioconductor class, namely a SummarizedExperiment.

  1. We could move everything over to S4. This would mean making moanin_model a S4 class, and the existing functions into S4 methods.
  2. We could just have a catch inside all of our functions that would test if data inherits from a SummarizedExperiment class, and extract the data from there. Note that SummarizedExperiment can hold multiple assays, and we'd have to use the default assay. The technical reviewers might complain about this.
epurdom commented 4 years ago

Also, if working with SummarizedExperiment, could assume that the relevant meta data is saved in colData(x), so can make a moanin_model with that information. If move to a S4 class, could actually extend the SummarizedExperiment class so that data and meta data are kept together in the object.

epurdom commented 4 years ago

Working on this on branch EAP/moaninClass.

NelleV commented 4 years ago

Can this ticket be closed? I think all the relevant code has now been merged into master.