Difference between revisions of "Annotation camp report"

From GO Wiki
Jump to navigation Jump to search
Line 1: Line 1:
GO Annotation Camp
GO Annotation Camp


The 3rd GO annotation camp was held at Stanford University on July10-14 2006. During the first two days members of the GO consortium met to discuss annotation policy; the main topic was evidence code usage. Ideally, all databases should use evidences codes in the same way but it has become clear that usage between groups can be inconsistent. In most cases, differences arise when there is a choice of legitimate evidence codes for a given experiment and different databases adopt their own preferences  e.g. a protein binding experiment may be annotated with IDA (inferred by direct assay) or IPI (inferred by physical interaction) evidence codes. During the discussion we were able to establish preferences for several such cases and these will be included in new evidence code guidelines.  In addition, the meeting agreed to refine the use of certain codes; in some cases usage has been extended and in others conditions for use have made more stringent. All of the recommendations will be incorporated into revised evidence code documentation that will clarify the use of codes and should improve annotation consistency between groups. Look out for the new documentation on the GO web soon. 
The 3rd GO Annotation Camp was held at Stanford University on July 10-14 2006. During the first two days members of the GO consortium met to discuss annotation standards,focusing on the consistent use of evidence codes. During the discussion we were able to clarify usage guidelines for cases in which multiple evidences codes could be used to support a GO annotation (e.g.a protein binding experiment may be annotated with IDA (Inferred by Direct Assay) or IPI (Inferred by Physical Interaction) evidence codes. In addition, we agreed to refine the use of certain codes or extend the use of other codes in light of new experimental approaches. All of the recommendations will be incorporated into revised evidence code documentation (http://www.geneontology.org/GO.evidence.shtml) that will clarify the use of codes for new groups and should improve annotation consistency between current groups.


The second part of the meeting was devoted to training. Around 30 curators, representing a diverse mix of plant, animal, fungal and bacterial databases, were introduced to various aspects of GO in a series of presentations by GO consortium members. There was also plenty opportunity to gain practical experience by annotating a representative set of ten publications; first working together in small groups guided by an experienced GO curator and then agreeing a consensus set of annotations as a large group. This format worked well and stimulated lots of discussion. This set of reference papers, annotated under the new guidelines, represents a valuable training resource that will be available on the GO web site soon.  
The second part of the meeting was devoted to training 30 curators how to manually curate GO annotations from literature. These curators represented a diverse mix of plant, animal, fungal and bacterial databases. The training included presentations about the onotologies, the evidence codes and the process of annotation, by GO consortium members as well as the opportunity to gain practical experience by annotating a representative set of ten publications in small groups guided by an experienced GO curator. This set of reference papers, annotated under the new guidelines, represents a valuable training resource that will be available on the GO web site.  


Thanks go to all at SGD in Stanford for arranging a great meeting. If you are interested in attending next year's annotation camp, please contact gohelp@genome.stanford.edu.
Thanks to Incyte and the Department of Genetics at the School of Medicine, Stanford University for supporting this meeting.If you are interested in attending next year's annotation camp, please contact gohelp@genome.stanford.edu.

Revision as of 05:29, 4 August 2006

GO Annotation Camp

The 3rd GO Annotation Camp was held at Stanford University on July 10-14 2006. During the first two days members of the GO consortium met to discuss annotation standards,focusing on the consistent use of evidence codes. During the discussion we were able to clarify usage guidelines for cases in which multiple evidences codes could be used to support a GO annotation (e.g.a protein binding experiment may be annotated with IDA (Inferred by Direct Assay) or IPI (Inferred by Physical Interaction) evidence codes. In addition, we agreed to refine the use of certain codes or extend the use of other codes in light of new experimental approaches. All of the recommendations will be incorporated into revised evidence code documentation (http://www.geneontology.org/GO.evidence.shtml) that will clarify the use of codes for new groups and should improve annotation consistency between current groups.

The second part of the meeting was devoted to training 30 curators how to manually curate GO annotations from literature. These curators represented a diverse mix of plant, animal, fungal and bacterial databases. The training included presentations about the onotologies, the evidence codes and the process of annotation, by GO consortium members as well as the opportunity to gain practical experience by annotating a representative set of ten publications in small groups guided by an experienced GO curator. This set of reference papers, annotated under the new guidelines, represents a valuable training resource that will be available on the GO web site.

Thanks to Incyte and the Department of Genetics at the School of Medicine, Stanford University for supporting this meeting.If you are interested in attending next year's annotation camp, please contact gohelp@genome.stanford.edu.