specificity_rank has NaN values from rank_aggregate()

[AtlasMaroc]

hello, I am trying to run rank_aggregate() but the column for specificity_rank is just NaN valuesc,

aggregate_ccc = li.mt.rank_aggregate( adata_subset, groupby = 'cell_type', resource= resource , #defined from mouse_consensus inplace= True, key_added= 'mouse_consensus_interactions' )

[dbdimitrov]

Hi @AtlasMaroc,

My guess would be that maybe you are passing z-transformed counts (stored in .raw) and these result in a NaN in logfc (with a warning?) and then also in specificity_rank.