Closed corneliusroemer closed 1 year ago
Thank you for asking about this, @corneliusroemer. There's no good reason to not upload sequences.fasta and metadata.tsv now. The only reason I didn't originally was because I wasn't sure whether the S3 uploads would be useful or how they would work. Now that they are working, we can upload the more useful downstream files.
You already upload raw fasta to AWS in
rule upload_sequences
.It would be useful for our https://github.com/neherlab/flu_frequencies repo to be able to download metadata.tsv as produced by your
rule parse
.I will duplicate your parse for now, but in the future it would be nice to reduce code duplication and parse in one place.
Is there a reason you are currently only uploading the raw data not the parsed one?