Hey!
mia has accumulated many datasets over the years. In addition, they are scattered between mia, miaTime and miaViz. In the long run, it might be better to keep those datasets in a separate data package (miaData?). This repository already provides a template, so we could tidy it up and turn into a proper data package for mia.
A few action points:
- adopt standard structure for data packages (e.g., ariadneDB)
- keep only RDS objects and store raw data elsewhere (e.g., Zenodo)
- (optional) add rds objects to ExperimentHub to be able to download individual datasets (there might be other ways)
- when ready, submit to Bioconductor
Hey!
mia has accumulated many datasets over the years. In addition, they are scattered between mia, miaTime and miaViz. In the long run, it might be better to keep those datasets in a separate data package (miaData?). This repository already provides a template, so we could tidy it up and turn into a proper data package for mia.
A few action points: