search

saezlab / liana-py

LIANA+: an all-in-one framework for cell-cell communication
http://liana-py.readthedocs.io/
GNU General Public License v3.0
166 stars 21 forks source link

li.multi.df_to_lr function gave gene Value error #67

Closed Hweeseon closed 1 year ago

Hweeseon commented 1 year ago

Hi ! I was trying "Hypothesis-testing for CCC & Downstream Signalling Networks" tutorial.

During the "DEA to Ligand-Receptor Interactions" part I confront an issue.

When I run,

lr_res = li.multi.df_to_lr(aadata, dea_df=dea_df, resource_name='consensus', expr_prop=0.1, # calculated for adata as passed - used to filter interactions groupby='cell_type2', stat_keys=['stat', 'pvalue', 'padj'], use_raw=False, complex_col='stat', # NOTE: we use the Wald Stat to deal with complexes verbose=True, return_all_lrs=False, )

I get warning that "all gene name values are not are not valid obs/ var names or indices." I checked, that aadata's var names has all genes. I am attaching a screenshot of the situation.

aaaaa bbbbb

I wonder what went wrong.

Best regards

Hweeseon commented 1 year ago

I'm attaching a screenshot of dea_df and aadata.var as well.

asdasd

dbdimitrov commented 1 year ago

Hey @Hweeseon,

Is it possible to upload a small subset of your data? I'm happy to debug it, as I can't see what could be an issue here. Besides, maybe there being no overlap between the resource (I.e. ligand and receptor genes) and your .var?

Hweeseon commented 1 year ago

Okay. I'll send subset data for this issue via e-mail to you.

dbdimitrov commented 1 year ago

@Hweeseon I will close the issue as it seemed to be related to the data itself as discussed via email.

Feel free to open new ones :)