Oct 2012 Meeting to finalize GPAD specification (Archived)
Link to previous proposed GPI format; http://wiki.geneontology.org/index.php/Gene_Product_Data_File_Format
and previous proposed GPAD format; http://wiki.geneontology.org/index.php/Gene_Product_Association_Data_%28GPAD%29_Format
Issues for discussion:
gp_association files
Questions awaiting discussion with Chris;
- Translating existing qualifiers CONTRIBUTES_TO and COLOCALIZES_WITH into annotation relations (+ NOT?)
- Filling in relationships retrospectively - implicit or explicit values in qualifier column?
Proposed format (21 Nov 2012)
contents | required? | cardinality | old column # | extra info |
---|---|---|---|---|
DB | required | 1 | 1 | must be in xrf_abbs |
DB_Object_ID | required | 1 | 2 | canonical or spliceform ID |
Qualifier | optional | 0 or greater | 4 | (NOT or integral_to)? (other_organism or colocalizes_with or contributes_to)? annotation_relation |
GO ID | required | 1 | 5 | must be extant GO ID |
DB:Reference(s) | required | 1 or greater | 6 | DB must be in xrf_abbs |
Evidence code | required | 1 | 7 | from ECO |
With (or) From | optional | 0 or greater | 8 | |
Interacting taxon ID (for multi-organism processes) | optional | 0 or 1 | 13 | ncbi taxon ID |
Date | required | 1 | 14 | YYYYMMDD |
Assigned_by | required | 1 | 15 | from xrf_abbs |
Annotation Extension | optional | 0 or greater | 16 |
gp_information files
1. Have decided columns for DB_subset (e.g. Swiss-Prot, TrEMBL) and Annotation Target Set (e.g. Reference Genome) are not necessary.
2. Includes a column for Xrefs to other databases (Col. 9). This would be useful for mapping of MOD-specific identifiers/symbols/synonyms to UniProt accessions to assist MOD curators moving to Protein2GO in searching for familiar IDs/gene names.
3. It is proposed to dispose of the 'DB' column (e.g. UniProt, WB) and replace it with a header line specifying the namespace of the annotating groups' identifiers, e.g. WB:, UniProtKB:?
None of the gp2protein files in the GOC SVN repository refer to objects from more than one namespace in column 1, so it seems like unnecessary repetition and redundancy to repeat the namespace in every row of the gpi file. Parent_object_id would only need to be qualified if it referred to an object from a different namespace (this could happen if we start annotating protein complexes)
Proposed format (21 Nov 2012)
column | name | required? | cardinality | GAF column | Example for UniProt | Example for WormBase |
---|---|---|---|---|---|---|
01 | DB_Object_ID | required | 1 | 2 | Q4VCS5-1 | WBGene00000035 |
02 | DB_Object_Symbol | required | 1 | 3 | AMOT | ace-1 |
03 | DB_Object_Name | optional | 0 or 1 | 10 | Angiomotin | |
04 | DB_Object_Synonym(s) | optional | 0 or greater | 11 | KIAA1071|AMOT | ACE1 |
05 | DB_Object_Type | required | 1 | 12 | protein | gene |
06 | Taxon | required | 1 | 13 | taxon:9606 | taxon:6239 |
07 | Annotation_Completed | optional | 1 | - | 20120614 (YYYYMMDD) | 20100405 |
08 | Parent_Object_ID | optional | 0 or 1 | - | Q4VCS5 | WBGene00000035 |
09 | DB_Xref(s) | optional | 0 or greater | - | - | UniProtKB:P38433 |