Review database cross-references

[eparejatobes]

Right now there are 155 database cross-references in UniProt:

• http://www.uniprot.org/database/

I want to discuss here

• [ ] which ones are essential
• [ ] a reasonable, flexible model for cross-references

[eparejatobes]

The idea is fairly simple:

1. If a cross-reference is deemed useful enough to be imported, that implies that the database itself to which it points is a useful resource. Then, it should be modeled and included at some point
2. Normally these cross-references point to something that would be a vertex, so we just create a graph for that DB and add vertices for the linked entities, having only their IDs

[rtobes]

This is the selection of the most important cross-references:

• http://www.uniprot.org/database/?sort=&desc=&compress=no&query=id:5%20OR%20id:16%20OR%20id:22%20OR%20id:23%20OR%20id:147%20OR%20id:148%20OR%20id:149%20OR%20id:150%20OR%20id:151%20OR%20id:24%20OR%20id:165%20OR%20id:141%20OR%20id:170%20OR%20id:28%20OR%20id:37%20OR%20id:51%20OR%20id:52%20OR%20id:53%20OR%20id:117

[rtobes]

Ranked by importance:

The most important:

    GO

Gene Ontology · UniProtKB (42,738,622) Category: Ontologies

   InterPro

Integrated resource of protein families, domains and functional sites · UniProtKB (51,058,113) Category: Family and domain databases

   ENZYME

Enzyme nomenclature database · Category: Enzyme and pathway databases

Sequence databases:

   EMBL

EMBL nucleotide sequence database · UniProtKB (64,324,983) Category: Sequence databases

   GenBank

GenBank nucleotide sequence database · Category: Sequence databases

   RefSeq

NCBI Reference Sequences · UniProtKB (33,494,473) Category: Sequence databases

Gene and genome annotation databases:

Ensembl

Ensembl eukaryotic genome annotation project · UniProtKB (1,231,040) Category: Genome annotation databases

EnsemblBacteria

Ensembl bacterial and archaeal genome annotation project · UniProtKB (29,644,899) Category: Genome annotation databases

EnsemblFungi

Ensembl fungal genome annotation project · UniProtKB (5,116,290) Category: Genome annotation databases

EnsemblMetazoa

Ensembl metazoan genome annotation project · UniProtKB (1,043,430) Category: Genome annotation databases

EnsemblPlants

Ensembl plant genome annotation project · UniProtKB (1,446,406) Category: Genome annotation databases

EnsemblProtists

Ensembl protists genome annotation project · UniProtKB (1,538,599) Category: Genome annotation databases

   PATRIC

Pathosystems Resource Integration Center (PATRIC) · UniProtKB (5,907,286) Category: Genome annotation databases

  BioCyc

BioCyc Collection of Pathway/Genome Databases · UniProtKB (4,685,844) Category: Enzyme and pathway databases

Networks, interactions, pathways, metabolism:

STRING

STRING: functional protein association networks · UniProtKB (7,617,500) Category: Protein-protein interaction databases

UniPathway

UniPathway: a resource for the exploration and annotation of metabolic pathways · UniProtKB (3,221,066) Category: Enzyme and pathway databases

IntAct

Protein interaction database and analysis system · UniProtKB (60,278) Category: Protein-protein interaction databases

   DIP

Database of interacting proteins · UniProtKB (20,104) Category: Protein-protein interaction databases

 KEGG

KEGG: Kyoto Encyclopedia of Genes and Genomes · UniProtKB (12,259,742) Category: Genome annotation databases

[rtobes]

I know that Biosystems is not a cross-reference of Uniprot but I think that in the future we could add this database to Bio4j:

• http://www.ncbi.nlm.nih.gov/Structure/biosystems/docs/biosystems_about.html

Theoretically it is open and it has a FTP to download its data:

• ftp://ftp.ncbi.nih.gov/pub/biosystems/

I have worked with this database and it is useful, it is updated, cover all the organisms and cover metabolic and signalling pathways.

The connection with Uniprot could be:

• via the biosystems NCBI protein ID that I think is a cross-reference of Uniprot (It is needed to check ifit is the same ID type in Uniprot and in Biosystems)
• via gene ID that I think is a cross-reference of Uniprot (It is needed to check ifit is the same ID type in Uniprot and in Biosystems)
• via taxonomy ID (In this case we sure that we have the NCBI taxonomy in both databases)