Jimmy O'Donnell
Jimmy O'Donnell
Weird. Is this the first time using the hashing option for dereplication? Or has it been used successfully, and now it's breaking?
It doesn't look like it's truncating because of length; it just doesn't include `;size=***`. When I run `bash banzai.sh test` without hashing, I get a dups_to_otus.csv that looks like this:...
The file `OTU.map` looks as expected for both unhashed and hashed sequence IDs for me (below). I take it you're using another version of at least some component of this?...
Yeah, since we have only ever used data with two levels of indexes, I never expanded functionality to accommodate data with only a single level of indices. I can work...
To date, I have dealt with this by hand on a per-issue basis, just to avoid additional changes to banzai.
Salut @fmichonneau, I have made this change but only incorporated it into the branch `dev`. It will be folded into master soon. In the meantime, if you move to the...
Hey @fmichonneau, if you ever tested this, did you find the example and description that I added to be sufficient? Happy to take more input. I think in the end...
It's doable, but... I kept away from it early on because changes were so frequent and I was less comfortable committing to a version numbering system. Even still, I wonder...
Yeah, I'd wanted to do that from the get-go, but there is no method I'm aware of to embed the date that it was downloaded from github. When you say...