Marco Galardini

Hi @flashton2003, what I can see so far is a slight difference in the number of input kmers: ``` 16630178 loaded variants 1028222 filtered variants 15601956 tested variants 15601956 printed...

Hi @flashton2003, could you clone the repository, activate your environment and try to run the unit tests? That could help us diagnose the problem or its source a bit better....

Hi @flashton2003, maybe pytest is not the best idea to make sure you are running the tests correctly; from the repository's root try `python setup.py test`. I think the best...

Seems like a great idea!

"they're coming outta the goddamn walls!"

Hi, yes, it is true that the only way to make this work is to flip any random phenotype. Maybe we can add a warning message when something like this...

I see! I'll try [setting `maxit`](https://github.com/johnlees/glmnet_python/blob/master/glmnet_python/glmnetSet.py#L208) to a larger number when calling `cvglmnet` and see what happens. If that is really the culprit we could use a larger default either...

Seems like a good idea! An alternative option could be to have a script that does the conversion, so pyseer doesn't need to change.

Hi, thanks for reporting this; I think this warning has been [introduced since python 3.8](https://bugs.python.org/issue32236). I've checked and this warning is raised whenever `pybedtools` is called. I'll check to see...

This is the issue opened for `pybedtools`: https://github.com/daler/pybedtools/issues/322