pl.ligrec does not match pvaule table output #1201

Report

Hi all,

I'm still new to bioinformatics and coding, having the following problem:
First of all I was wondering how it is possible that pvalues can be exactly 0.
Based on this observation, the plot of the interactions seems to be off. I noticed for one interaction pair that the values in the pvalues table are either 0 or NaN. In the plot however the dots for the 0 values are marked as not significant. Thanks for your help in advance :)

I used the following code:

res = sq.gr.ligrec( adata, cluster_key="celltype", n_perms=500, threshold=0.05, copy=True, use_raw=True )

sq.pl.ligrec( res, alpha=0.05, swap_axes=False, means_range=(1, np.inf), target_groups="C LAM" )

Versions