Database maintenance workflow (curated genomes, reps, pangenome ranks, merged) #3782
Description
Pull request that @bluegenes and I have been working on and off over the last year to make database generation much easier.
sourmash-bio/database-releases#6
This workflow creates a database from the current assembly_summary NCBI information. This includes removing the redacted genome assemblies while updated the genomes to the current version in NCBI.
There is a workflow for updating GTDB and Genbank domain databases with a config file for workflow variables.
Do you think it would be worth the diskspace for there to be databases for:
- genomes
- reps
- species
- genus
- family
- everything merged
I've built all the ranktables for gtdb at a species level and I think there should be an easy way of concatenating the species level ranktables into any high rank so there is no need to duplicate them. That said, should I create a ranktable directory for every species database to have existing for easy comparisons?