Ideas for publicizing Ref.Genome Annotation Data (Retired)
How is your group Publicizing Ref.Genome Data
- GOA: GOA currently publicizes the RefGen project here: http://www.ebi.ac.uk/GOA/RGI/index.html. Additionally, GOA users can also filter for Reference Genomes annotations for human and chicken taxons in QuickGO, with this URL:
- BHF-UCL: The BHF-UCL team link to the GOC RefGenome page from our home page
- MGI: MGI create a "The Gene Ontology Consortium Reference Genome Annotation Project at MGI" page under our Functional Annotation home page with a link to the GOC RefGenome page and links to two files a)MGI genes selected for the project b)Comprehensively annotated MGI genes. http://www.informatics.jax.org/mgihome/GO/reference_genome_project.shtml. MGI also tag every single reference gene on the gene report/detail pages.
- RGD: Planning a web page describing the project with links to and from all appropriate gene report pages.
- TAIR: Info [page] on website. We would like to tag the ref genes but have not yet done this. Also, featured RefGenome in Fall 2009 newsletter.
- Wormbase: We wrote an item on the project in a past Newsletter. News item about this project is currently on the front page of WormBase in 'News and Notes' section. Reference genome genes are already being tagged in our internal curation database, with the idea that in the future we will also tag these on our web page.
- ZFIN: We discussed the project in a recent news letter. I have several work requests in to identify the Ref.Gen genes in ZFIN. We may suggest our users visit AmiGO (we would link there) to explore data on these genes further. How much we can do depends on how priorities are set for our developers for the coming year. I can't guarantee anything, but will continue to lobby for progressive change.
Ideas for Publicizing Ref.Genome Data
Reference Genome Annotation data should be made available at each of the participating groups (for download and display) and should be available in AmiGO in a user-friendly way.
At each MOD
- How can the MODs (ref.genome project member) highlight the genes that have been annotated for the ref.genome project?
- What do we want our users to get out of this annotation data?
- Is it useful to just say 'XXX1 is a reference genome gene'?
- What aspects of the ref.genome annotation data should be highlighted?
- Should all groups link out to one common resource (may be in AmiGO)?
feedback from GOA
- this is where GOA currently publicizes the RefGen project: http://www.ebi.ac.uk/GOA/RGI/index.html
- GOA users can also filter for Reference Genomes annotations for human and chicken taxons in QuickGO, with this URL: http://www.ebi.ac.uk/QuickGO/GAnnotation?&protein=ReferenceGenome
- Perhaps we could highlight the genes that are targeted by RefGen for the next month from the GO Consortium page - have a prominent link/short piece of text on the front page, that would link out to a list of the genes being targeted over the next month. If all MODs linked out to this text, then it would mean that only one person would need to regularly look into updating a page, rather than each MOD - meaning it would be more likely to be kept up-to-date and in sync with RefGen work. This would also mean that the GOC site, which must be a highly popular site with eager GO users, would more visibly promote the RefGen project.
Edimmer 09:48, 29 January 2010 (UTC)
AmiGO is the face of GOC and is good place to promote and offer useful displays for ref.genome data.
Here are some ideas. Please feel free to add.
- We should be able to download a list of genes picked for ref.genome project. In addition users could have an option to filter annotations by species
feedback from GOA
- It seems sacrilegious to say so, but why should AmiGO users be intrinsically interested in RefGen annotations in particular - as over the last couple of years we have annotated a broad range of (mostly) unconnected genes?
- However, now that RefGen targets are now being selected as components of a larger, more connected lists - such as the MGI lung branching project, perhaps AmiGO could provide a specific filter to display subsets of targets/annotations from specific RefGen annotation projects. Researchers from this domain could then be encouraged to feedback to the group on these annotation lists.
- Shouldn't some of the other annotation projects, which have won funding - such as the British Heart Foundation Cardiovascular gene set, and the Kidney Research UK gene set be highlighted in AmiGO in a similar way? There are 3 full-time GO curators who are specifically funded to providing these focused annotation resources. Their annotations are very valuable resources for the cardiovascular/renal communities. If a list of targetted annotation projects could be generated (including RefGen Lung branching, then these projects should also be listed)
- I think that a tag(star or dendrogram) next to each RefGen targeted gene to indicate that there had been a focused curation effort, and encouraging feedback should be enough?
- Annotations are created by homology-based methods by many different groups, PAINT is just one of these (although it should become a very powerful provider of annotations).PAINT-provided annotations should, like all annotation methods, fully describe the annotations via its GO_REF - and link out to pages that fully document the tool/resource. I like SGD's the idea of linking out to annotations that have been created by homology methods from the source annotation - but this should be provided for all homology-based annotation methods that groups provide the GOC.
Edimmer 10:07, 29 January 2010 (UTC)
Some feed back from SGD curators
1) Turns out some of the SGD curators did not realize that ref.genome data was integrated within AmiGO. I strongly recommend that other MODs also bring this up within their group to educate and garner ideas.
2) How to look for ref.genome data in AmiGO? - there is no link on the home page and since ref.genome data is a resource and not a tool, one wouldn't think of clicking on the More tools link.
2a) Perhaps remove this option from the Tools tab and have a separate tab called 'Homolog Annotations' (or somthing that we can agree on. We shdn't use ref.genome as a tab name because it is too jargony and no body will click on it)?
2b) When you click on the "More tools" tab, you get to a page that is titled- "Complete List of AmiGO Software" The word software is too jargony. how about change it to just "GO Tools"? BTW, BLAST should *also* be listed under Tools.
3) If I search for TOP1 (with the genes/proteins radiobutton checked), I get a list of TOP1 genes and the ref.genome ones have a 'star' next to it. But the star is totally lost on the page (very small and not bright enough). We should look into highlighting those with gold stars (good size and bright) or some thing else nice and bright (may be a dendogram icon)
4) Once you click on say the S. cerevisiae TOP1 from the page I mentioned above, you get to the TOP1 details page. http://amigo.geneontology.org/cgi-bin/amigo/gp-details.cgi?gp=SGD:S000005366&session_id=9034amigo1263416858
4a) We all felt that the phrase 'Ref Genome' and then 'Homology under TOP1' is not clear, is confusing and we should come up with some intuitive language to highlight what that link is about. There are two different links in that one line, one going to the annotations of TOP1 homologs and one going to the complete homolog set (list of genes). They should be separated and reworded.
For the Annotations- the link should be reworded to something like (of course other groups shd be consulted on the wording):
- Annotations for homologs [tab] TOP1 related genes in M.musculus, S. cerevisiae ..etc (keep the colored icons)
hyperlink 'TOP1 related genes' OR
- Annotations for homologs [tab] Genes similar to TOP1 in M.musculus, S. cerevisiae ..etc
hyperlink 'Genes similar to TOP1' OR just say in one long line-
- Compare GO annotations for TOP1 and its homologs in M.musculus, S. cerevisiae ..etc (keep the colored icons)
hyperlink "Compare GO annotations for TOP1 and its homologs"
4b) The 'Homology under TOP1' link should possibly be changed to 'List of TOP1 related genes in model organisms'. We could probably use the gold star or what ever icon we decide on this page and consistently wherever applicable to indicate ref.genome project.
5) Ref.Genome genes are not currently highlighted when you come in from a term search. For example when you look at the list of gps associated with say GO:3917 (DNA topoisomerase type I activity), the ref.genome genes are not highlighted with a star on this list. As I mentioned earlier, we should highlight ref.genome genes with a star, wherever it is displayed.
- Should we indicate ref.genome genes with a Y/N column in the GAF file?
feedback from GOA
- I agree with SGD that we should indicate refgen genes in a better way than we currently do. We don't have a good way of indicating this for internal or external purposes.
- Perhaps the gene product information file that Amelia was suggesting might be a better place than the GAF however? As the GAF is centered around annotations whereas the gene product information file would be centered around providing information on the genes/proteins. See:http://wiki.geneontology.org/index.php/Annotation_File_Format_Proposal (I've also added this suggestion here as well)
Edimmer 10:14, 29 January 2010 (UTC)