Re: bind results

Helena,
  the BIND endpoint refers to the BIND (Biomolecular Interaction Network
Database) database, not BindingDB.

we should note that the BIND db is now deprecated, as we are parsing
iRefIndex for all PPI from  BIND, BioGRID, CORUM, DIP, HPRD, InnateDB,
IntAct, MatrixDB, MINT, MPact, MPIDB, MPPI and OPHID.

http://irefindex.uio.no/wiki/iRefIndex
http://cu.irefindex.bio2rdf.org/sparql/

m.

On Wed, Jun 13, 2012 at 9:29 AM, Helena Deus <helenadeus@gmail.com> wrote:

> Hi there,
>
> I am running the following query in BIND through federated query:
>
> select *
> where {
>
> SERVICE <http://bind.bio2rdf.org/sparql> {
> ?protein1 <http://bio2rdf.org/ns/bio2rdf#xGI> ?protein2 .
> ?protein1 a <http://bio2rdf.org/ns/bind:MolType-protein> .
>  ?protein2  a <http://bio2rdf.org/ns/bind:MolType-protein> .
> filter ( ?protein1 !=  ?protein2 )
> }
>
> }
>
> This strange list of "protein associations" does not seem to make sense;
>
> Some of these associations, e.g.
> <http://bio2rdf.org/gi:9606> <http://bio2rdf.org/gi:33589859>
>
> Appear to come from the BIND raw file "BDB_cid.txt <http://www.bindingdb.org/bind/BDB_cid.txt>(
> 5.72 MB ) Mapping between BindingDB monomerid and PubChem cid"
> Which would mean that gi:9609 is a monomer and gi:33589859 is a pubchem
> cid, which is a compound (in this case, it goes by the very long name of tetra-O-benzyl-d-mannitol
> deriv. 1jj, D-Mannitol,
> 1,6-bis-O-[(2-fluorophenyl)methyl]-2,5-bis-O-[(4-fluorophenyl)methyl]-)
>
> So, I think that a whole bunch of identifiers appear to be
> misclassified... Unless I am missing something...
>
> Any help with be greatly appreciated!
> Best,
> --
> Helena F. Deus
> Post-Doctoral Researcher at DERI/NUIG
> http://lenadeus.info/
>
>


-- 
Michel Dumontier
Associate Professor of Bioinformatics, Carleton University
Chair, W3C Semantic Web for Health Care and the Life Sciences Interest Group
http://dumontierlab.com