Adds crested.get_dataset(), currently with support for BICCN topic bed and peak bigwig datasets. Example dataset names (currently "mouse_cortex_bed"/"mouse_cortex_bigwig") can still be changed, but do remember to change introduction.ipynb as well.
Open questions/to-do:
[ ] Add melanoma and fly brain data
[ ] Add DARs for transfer learning example (on whichever dataset we end up showing that)
[ ] Should we refer to a tutorial to preprocess data like this?
If so, should it be to get these specific files (esp for topics: BICCN with this topic modeling and otsu cutoffs), or is referring to pycisTopic/snapATAC2 tutorials enough?
I'm already merging this since there are no conflicts anyway and I want to test with new main branch functionality.
Feel free to continue working on this in this branch and create a new PR later.
Adds
crested.get_dataset()
, currently with support for BICCN topic bed and peak bigwig datasets. Example dataset names (currently"mouse_cortex_bed"
/"mouse_cortex_bigwig"
) can still be changed, but do remember to changeintroduction.ipynb
as well.Open questions/to-do: