All p.val are N.A. for mixedmodel

[DelongZHOU]

Hello,
I'm running the DE with a mixed model as following:

de_family = 'mixedmodel', de_method = 'negbinom_offset', de_type = 'LRT',

The script is completed without errors. However all the p value and the adjusted p values are NA for every comparison, for my two datasets.

Is this expected behavior? I saw in previous issues other people had mixed value / NA for their p val when running the mixed model, but apprently they were using an older version where the family in the output file was pseudobulk. Mine is correctly marked as mixedmodel.

Please let me know if there's any other information I can provide to identify the issue.

Thanks,

[DelongZHOU]

Hi,
Just for the fun of it I also ran negbinom without offset, as well as poisson with off set.
Both combinations also gave NA for all the p val.

[DelongZHOU]

Alan has pushed a fix for this problem. It seems that the labels were not passed correctly when using the cell_type_col, replicate_col, and label_col arguments in the run_de().