GO slim overhaul (completed 2009)

The generic GO slim is badly in need of an update. At present this page is just a place to note ideas ...

Personnel: Jane, Val, ??

• Ideally want three "generic" slims:
  • truly generic slim for all species (probably very high level indeed)
  • slim for multi-cellular organisms
  • slim for single-celled organisms
• could also do eukaryotic vs. prokaryotic

• Other considerations
  • Email with specific queries about terms included and omitted from current (as of June 2009 page creation, actual last update much earlier)
  • Mailing list thread with Val's criteria for term selection
  • Can we avoid "other X" terms? They're confusing and hard to handle.
  • Try to make slims is_a complete

Tangentially related: Does anyone know of a web-based slimming tool which shows the number of gene products which are annotated, but not to any term in your slim, and the number which are not annotated to any GO term (i.e root node annotations)? (question from Val)

Linda suggested: 1. slim for metagenomics 2. Collecting purpose made slims

Val Some possible criteria which have been suggested previously that we may wish to consider http://fafner.stanford.edu/pipermail/go/2008-May/015997.html

Amelia: The hardest part of automatation will be to identify "biologically relevent terms". may be necessary to look at "extermal sources of terms"