#TODO
-
magine_analysis.py
-
Restructure class to have setters for each data type.
-
Make pandas dataframe get lists per sample id, get rid of return_rna, return_proteomics
-
Have a sorting method to sort sample_ids, right now we assume they can be floats but if they are strings they should be able to defined the order.
-
-
Function requests
-
Add -log10 pvalue and log2 fold change to network node attributes.
-
For network subgraphs, collapse all unmeasured nodes into the edges.
-