refchef-view and standardized metadata should follow genbank metadata

[aguang]

I am proposing that metadata for refchef and refchef-view should follow ones for GenBank. In particular, we should have the field Organism (instead of species) and GenBank assembly accession. We could also enforce a RefSeq assembly accession. The GenBank and RefSeq fields can be filled out as custom if it is a custom reference built with assemblers from NGS data, but we should have a standardized value for such custom references.

I believe the requirement to provide a GenBank assembly accession will make it easier for researchers to search for references and for us to mark duplicates in refchef requests as well. It will also make it easier to distinguish between different versions of references such as GRCh38/HG38 and hg19. Thoughts?

Also, 1) is the organization field where the genome was downloaded from or who submitted the genome? and 2) is there a mechanism in place to check or autocorrect for organism/species names? For example, do we want someone who enters human or Homo sapiens in the Organism field to be warned or have their entry autocorrected to Homo sapiens (human)?

[fernandogelin]

still need to update refchef-view with new fields.