Add GISAID isolate ID to acknowledgments download

[joverlee521]

Flagged internally by @rneher that the acknowledgements download does not include the GISAID isolate IDs.

#138 added the HA/NA segment accessions, but we do not include the isolate ID used to link multiple segments to the same isolate. The isolate id was added to the metadata upload in #160, so it should be available in the metadata for the builds.

The fields included in the acknowledgements is hardcoded in Auspice so the field will have to be named gisaid_epi_isl.

TODOs

• Rename metadata column from isolate_id to gisaid_epi_isl #179
• Re-run upload workflow to update S3 column names
• Use gisaid_epi_isl in workflows #180

[joverlee521]

@huddlej, would it break any of your workflows to rename the fasta_fields to gisaid_epi_isl?

[joverlee521]

Ah, at least the nextclade workflow is using the isolate_id field.

Maybe better to just rename isolate_id to gisaid_epi_isl within the phylogenetic workflow right before the augur export?

[huddlej]

I don't depend on isolate_id as a field name anywhere. Since Auspice hardcodes separate fields for GISAID and GenBank accessions (without an awareness of segmented viruses), it makes sense to rename isolate_id to gisaid_epi_isl in the uploaded metadata.

Unless the Nextclade somehow depends on an isolate_id field downstream, I would just change the Nextclade workflow to use the new field name, too.

[joverlee521]

Reopening because we've removed gisaid_epi_isl from metadata in #188.

This issue now depends on nextstrain/fauna#165