We see this happen when all of your samples (or, all but one?) are filtered out due to rarefaction. If you set the “even sampling depth” slider to 1000 in the feature-table summarize viz, how many samples remain?
I see what you are saying- I typed in 1000 and it gave me only 11% of my data, so that is probably one my issue. I was basing my sequencing depth number off my rarefaction curve that I had produced.
But I based my number on the maximum sequence coverage on the dada2 table but I still got the error message when I tried to create the core metrics:
tmp/q2-AlphaDiversityFormat-kqlgcdsq is not a(n) AlphaDiversityFormat file
Sorry, let me clarify - how many samples remain in the feature table after rarefying (not how many sequences). Here is an example screenshot from the viz I referenced above:
In this example there are 28 samples remaining at the selected sampling depth of 1347.