10 AUG 2010 RefGen Phone Conference (Archived)

From GO Wiki
Jump to navigation Jump to search

Present

wnt project

WNTs have cytokine activity [Ruth/Alex]

  • see PMID:18986540
  • Wnts identified as affecting the formation of synapses is controlling an effector function of cells.
  • Note due to lack of information provided in this paper only Wnt5a (R&D Systems only sells mouse Wnt5a) and Wnt7a (R&D Systems only sells human Wnt7a) can be annotated.

Subfamilies annotation report from Mary Dolan

I have put together a report (subfamilies.xlsx posted ftp://ftp.informatics.jax.org/pub/curatorwork/Pascale/) that summarizes the GO annotation status (as of August 1,2010) of Panther subfamilies according to the type of annotation for each refG organism. This is similar to something I had done for the old PPOD clusters.

There are separate sheets for function, process, and component. For each subfamily that contains a refG organism member, I display the type (color-coded) of annotation: experimental (green), ISS-type (yellow), other (red), or none (an organism entry for the subfamily indicates that it has a subfamily member, the slot will be empty if the organism has none). I include ICs as experimental annotations; ND annotations are not counted. I have added a few mappings from panther ids to ga file ids for completeness.

The numbers on the 'annotation_counts' sheet should be useful information for the GO grant renewal in that they give an indication of how much data that is already available in MOD ga files from years of annotation effort might be used by other annotation groups via the refG/Panther project. I think the numbers are quite impressive.

For example, of the 69,566 panther subfamilies: 32,991 have one or more model organism members (14,583 have two or more refG members). Of these 32,991 refG member subfamilies: 24,265 have some GO function annotation to one or more organisms; 15,276 have either experimental or ISS annotation; and 11,653 have experimental annotation. The numbers are similar for process and component.

I see this as complementary to the careful group by group literature curation followed by PAINT review, which, of course, provides the best comprehensive annotation. This provides a view of the maximum possible gain of shared information we could get for 'free' for other refG MODs or other genome annotation groups for their organism's Panther subfamily genes.

The report also has sheets with various counts of things -- e.g. how many genes for each organism in each subfamily. I have also done this report for panther families but I think the idea of shared information might apply more to subfamilies.

Hoping this will be useful, I look forward to your comments. Thanks. Mary

Mary E. Dolan, Ph.D.

Action items from last call

http://wiki.geneontology.org/index.php/13_JULY_2010_RefGen_Phone_Conference