From GO Wiki
We talked at the start of the grant period about producing quarterly reports - these would be for NIH and for ourselves. Should we start doing this? May also want to cross-reference with the grant milestones.
- Last summer, we surveyed groups on what they put in the With/From column for different evidence codes.
- We need to finalize some decisions about the syntax of With/From, specifically with respect to placing multiple values in With/From.
- IMP - some groups enter IDs for allelic variations and RNAi experiments. Multiple, pipe-separated values have been entered and are meant to indicate that the BP is inferred from each perturbation independently. If more than one variation within the same locus resulted in a phenotype, would those variations be comma-separated?
Two different deletion mutations and one RNAi inactivation support the same GO annotation Pipe-separate With/From for this annotation: WB:WBVariation00091989|WB:WBVar00249869|WB:WBRNAi00084583
- IGI - gene or gene product entities are typically entered in With/From for IGI. Multiple, pipe-separated values have been entered to indicate triple mutants, but should these values instead be comma-separated? Could pipe-separated values be used to indicate individual genetic interactions that result in the same inference for a process? This would save space in the GAF/GPAD and would be more consistent with how we use pipes and commas for annotation extensions.
With/From representation of triple mutant (currently): WBGene00000035|WBGene00000036 Possible new representation for triple mutant: WBGene00000035,WBGene00000036 Pipe-separated would then indicate independent interactions that result in the same GO annotation
- IEA - multiple, pipe-separated InterPro accessions are currently used for IEA-based annotations.
- We should consider whether we can treat the with/from fields for all of these in the same way, e.g. if you have lots of binding experiments in a paper, could you represent this with multiple with IDs separated by pipes? If a protein binds two proteins at the same time, would we want to represent this by putting the two IDs in the 'with' separated by a comma?
- Would we allow both pipes and commas in the same with field?
- Does this also extend to the taxon column?
Interaction between two species: taxon:6239, taxon:652611 Independent interactions used to annotate to the same GO term: taxon:6239, taxon:652611|taxon:6239, taxon:90371
- What downstream (consumer) effects would any changes on With/From and Taxon syntax have?
GOC Meeting Follow-Up
- Plans for coming year?