Stefan Peidli

This function at least subsamples all classes in an obs column to the same number of cells. Would be straightforward to modify to what you probably think of. ```py def...

Seems reasonable. No immediate problems this could cause come to my mind. And by having a function per metric we could add more docs including formulas, which I agree is...

In the meantime a small and dirty trick: `adata.var['chromosome'] = ['chr'+str(i) for i in adata.var['chromosome']]` Then it worked for me.

Note: `conda install -c anaconda pyqt` will lead to a version below the one required by pertpy (5.5.12 < 5.5.17).

Great suggestion! We usually calculate most distances in lower-dimensional spaces (such as PCA) since [distances in high dimensions are bad](https://en.wikipedia.org/wiki/Curse_of_dimensionality#Distance_function). Depending on how large your set of genes is you...

Ah sorry for overlooking this! Thanks @Zethson for pinging me! Honestly, I never calculated the variance with Wasserstein distance, so I do not know what sensible values for this case...