Metadata script does a lot of ENA API calls to get information for each genome. If ENA API is unstable - whole script fails -> pipeline fails. -resume function runs everything from scratch.
Maybe it makes sense to improve this script:
create a backup file for already fetched metadata
fetch as one big file and parse it
fetch metadata in the start of pipeline (in parallel with fetch ...)
Metadata script does a lot of ENA API calls to get information for each genome. If ENA API is unstable - whole script fails -> pipeline fails. -resume function runs everything from scratch. Maybe it makes sense to improve this script: