Does Ncbi Have A 'Best' Primary Key That I Can Link To?

Question

[David]

I have a database of plant phenotypes with a table of plant cultivars, varieties, and genotypes to species information in a local copy of the USDA plants database. In many cases, these records have cryptic identifications such as 'I-596', and the only information that I include in the table is the name and a citation (plus relevant growing location / treatments / conditions).

Other users of the database would find it useful to link these records to NCBI. However, it is not clear if I should link to a RefSeq, GenBank, both, or other. If so, which primary key(s) should I use (Accession number?)

Answer 1

[Niek De Klein]

If you only want to link it to NCBI I think the best thing is to link it to http://www.ncbi.nlm.nih.gov/gquery/ using the GI as primary key.

[David]

@Niek thank you for the answer. Your link does not provide easy access to more information about GI, although I found a place to start here: http://www.ncbi.nlm.nih.gov/Sitemap/sequenceIDs.html

[Niek De Klein]

The GI is the NCBI database accession number. Since you didn't specify what information the other users would be interested in, going to that link using the GI you can get all the NCBI information that is available.

[David]

@Niek thank you for your help. I don't know what information we will need (this is for unidentified potential users), but want to be able to provide a relevant link to NCBI if the information is available.