Hi @macrobiome,
Good! I always have to check 
I misunderstood how your metadata was set up. There is no issue with your donor metadata column.
However you will want to add a "fake" time column. I would put a column that is called week in your metadata. Give all the tumor values 1 and leave all the intestines blank. (This is part of the tweaking this method to fix your study)
Lastly remove the filtering parameter, you will need all your samples in order to run this anaysis.
--p-where '[SampleType]'="Tumor"'
Can you try those suggestions and let me know how it goes. If this doesn't work for you would you share your faith_pd_vector with me so that I can try it? (You can DM me your data if your want)
Hope that helps!
