Georges Kanaan
Georges Kanaan
@ArtRand would it make sense to concatenate one's samples into a single BAM and run `sample-probs` assuming there are not outlying samples?
You'll have to forgive my primitive understanding of the statistics here as I attempt to answer. My goal ultimately is to understand whether two tables are identical or not (we've...
Thank you for that clarification and suggestion. I've chosen to use the score test as I believe it to be more robust for my dataset with a small number of...
Thank you that's noted. I will run a wald test and compare results and report back with some more detailed information about how I ran them and results.
I tried to run a small example to test your suggestion and the Wald score, here's what I did: ``` X = pd.DataFrame({ 'name': [1, 1, 1, 2, 2], 'sample_0':...
@princi-pol I was wondering if you found an alternative solution? I also want to use MNLogit on a count table.
I used `0.3.1`, also the gene-coordinates file is the issue, just looked at it and it's not normal. Guess that was the issue! I'll fix it and confirm.
Seems like that fixed it @ArtRand next time I'll review my input files instead of trusting the script! Sneaky updates sneak pass me...
Hey art, I was wondering if you had a time line for this?
I may try to take a stab at this after ISME. I've written scripts to make custom HMM sources so I have those already, and wanted to look at viral...