You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I will list a few suggestions which might improve the usability of the package.
Allow to pass top_p for features in the heatmap
cmap parameter to heatmap (currently it's defined internally, so it cannot be passed to seaborn).
top_n parameter for plot_weights
allow key to be passed to plot_weights, e.g. key=abs would enable top absolute weights to be plotted.
Clarify in the documentation how are shared SVGs determined when concating anndatas from multiple samples, is it the union, intersect, or some proportion?
Color legend for plot_samples is it possible to also add a legend based on the ~most extreme archetype?
having adata.uns['chr_aa']['alphas'] in obsm instead of uns would make it much easier to index.
Again great work, and these are just suggestions :)
Daniel
EDITED
The text was updated successfully, but these errors were encountered:
Hi @rockdeme,
I will list a few suggestions which might improve the usability of the package.
top_p
for features in the heatmapplot_weights
plot_weights
, e.g.key=abs
would enable top absolute weights to be plotted.plot_samples
is it possible to also add a legend based on the ~most extreme archetype?adata.uns['chr_aa']['alphas']
in obsm instead of uns would make it much easier to index.Again great work, and these are just suggestions :)
Daniel
EDITED
The text was updated successfully, but these errors were encountered: