Bundles/ipi

PDB and some of our other data sources are indexed by Uniprot id, while the Neurocommons store only knows about Entrez Gene ids, so this is a bundle consisting of a Uniprot/Gene correspondence.

ipiparse.py parses data from a primary source:

It follows bio2rdf in naming uniprot ... thingies... um... records?

The following query shows the structure of the output:

Show Entrez gene of each PDB chain

prefix sc: <http://purl.org/science/owl/sciencecommons/>
prefix bio2rdf: <http://bio2rdf.org/ns/bio2rdf#>
select distinct ?chain ?sym  where 
{ 
  ?chain sc:uniprot ?prot.
  ?gene bio2rdf:xPath ?prot.
  ?gene sc:ggp_has_primary_symbol ?sym.
  filter ( regex(?sym, "^HLA-") )
}
order by ?chain


odd... the bio2rdf schema says http://bio2rdf.org/ns/bio2rdf:xPath but I'm sure I saw http://bio2rdf.org/ns/bio2rdf#xPath somewhere.