RefGenome10Feb09 Phone Conference (Archived)
Pascale, Mike, Kimberley, Stan, Kara, Mary, Donghui, Stacia, Emily, Rachael, Susan, Suzi, Petra, Doug, Ruth, Li, Judy, Chris, Varsha, Lakshmi, Harold
- hoping to have it available for this round of annotation
- pulling data from the GO database (annotations)
- need to implement the ability to write a GAF file
Next electronic jamboree Feb24
Please annotate genes !! Electronic_jamboree_feb24-2009
PAINT annotations: ISS
- PAINT annotations are done in two steps:
- Annotating the common ancestor (hypothetical protein) by ISS by 'tree annotators'
- This generates a GAF file that can be used to propagate those annotations to all other descendants of that sequence
- In both cases, the annotation uses ISS as evidence code. OK? YES
- Step 1 : annotate ancestral sequence with ISS with all sequences with EXP data
- Step 2 : annotate all other descendants with ISS with PANTHER ID
- Are they being used consistently and can we use them for propagation?
- Used with two different meanings: (1) when a multisubunit complex needs two or more gene products for activity, and (2) from documentation: "Note that contributes_to is not needed to annotate a catalytic subunit. Furthermore, contributes_to may be used for any non-catalytic subunit, whether the subunit is essential for the activity of the complex or not." 
- This last statement should be removed. It's incorrect to annotate a function as if annotating a component.
- contributes_to for RG genes (txt); File:Contributes to RG-2009-02.xls (xls)as of Feb 9, 2009:
- Harold: this should be used only when it's absolutely required
- sometimes get confused with regulation
- Doug: should we carry the contributes_to when doing ISS?
- Harold: for EXP: it depends on the experiment that was done
[ACTION ITEM]: fix documentation
- Used with two different meanings: (1) transient localization, and (2) peripheral association.
- Do we need to annotate peripheral associations? If the assay cannot resolve the component, should we not annotate to the parent?
- If we mean 'transient', should the qualifier be 'transiently localized' (ie, more explicitly saying what we're trying to capture?
- colocalizes_with for RG genes (txt); File:Colocalizes with RG-2009-02.xls (xls) as of Feb 9, 2009.
- Emily: problem with many people use them differently so we have all this various data to go back and fix
- Emily: should co-localizes be used for major terms such as leading edge, nucleus and cytoplasm (temporal sense)
[ACTION ITEM] (Emily): Send email to list to obsolete colocalises?? or remove the transient association meaning
[ACTION ITEM] (Pascale) generate lists of terms where it's been used
- From the TOP2 curation [] : Should Fraction terms be obsoleted? Since they do not represent cellular components? If people agree here, we can ask the GO/annotation lists.
- GO:0001950 : PME fraction => plasma membrane?
- GO:0000267 : cell fraction => cell?
- GO:0005625 : soluble fraction => intracellular?
- GO:0005624 : membrane fraction => membrane?
- GO:0005626 : insoluble fraction => membrane?
- GO:0042598 : vesicular fraction ??
- GO:0000299 : integral to membrane of membrane fraction => intrinsic to membrane ?
- GO:0000300 : peripheral to membrane of membrane fraction => membrane?
- GO:0005792 : microsome ??
GO:0019718 : rough microsome only few proteins, delete? GO:0019719 : smooth microsome only few proteins, delete?
Please see PMID: 12644571 for use of these GO terms
- Not comfortable deleting them yet; we would need to find a term to replace them. Cell part? Different kinds of cell parts?
- In some cases it's the only data available and does suggest direction of further experimentation. Replacing using above map scheme may be putting more interpretation into the experiment than the author was comfortable with.
Next conference call
Tuesday March 10, 2009, 11 AM PDT, 1 PM CDT, 2 PM EST, 7 PM BST
Return to Reference_Genome_Annotation_Project