Managers 16Dec09

From GO Wiki
Jump to navigation Jump to search

Participants:

Agenda: Jen

Minutes: David

Action items from previous calls

  • David, Tanya, Jane: Compose wiki documentation for ontology editors, with step-by-step instructions on making cross-products - in progress (will start this week)
  • Jen: Draft wiki page to summarize motivation for MF-BP links - in progress
  • Managers: Comment on draft proposal that Suzi will circulate (content meetings, RefGen annotations, etc.) - not sent
  • Pascale: Confirm date of Geneva annotation camp - 14th-17th June at SIB - sent to GO-Top and only Judy responded. GO-Top to confirm.
  • Midori, Jane, David, Tanya, Chris: Midori to resend draft (DONE), others to comment - sent - some comments received. In progress.
  • Chris filter IEAs from GAF inference. do we need more detailed evidence codes - e.g. IC-derived-from-ISS - in progress
  • Chris report on aligning Reactome and GO - in progress
  • Emily: We can use this text http://wiki.geneontology.org/index.php/GAF_2.0 to make documentation of GAF 2.0 for the website. Perhaps Emily and Amelia would like to work together on that.- in progess
  • Emily: Send a mail to GO-friends to advertise GAF 2.0 and to show the docs.- in progress

Discussion items

  • CHEBI XPs. The classification of some chemical entities such as "carbohydrate" in CHEBI is radically from GO (for example, in CHEBI, nucleotide is_a carbohydrate). Should we just adopt their classification wholesale? This would radically reorganize GO in places (Chris)

Right now, we seem to be at a bit of sticking point on this. It seems that CHEBI and GO want to have a different view of complex molecules. CHEBI wants a structural view and GO wants a metabolic/biochemical view. CHEBI classifies every carbohydrate-containing molecule a carbohydrate, but this causes problems when GO represents the metabolism/biosynthesis/catabolism of these molecules. We will continue to try to work with CHEBI to align our views.

  • UniProtKB IDs. (Chris)

It seems that the IDs have been split into SP and TrEMBL records. This is breaking some things in the pipeline. They did this to provide users with info about whether a record was a SP or TrEMBL record. Emily will check into this, but she feels that the UniProt users want to be able to distinguish records. Emily will work out a way to try to keep both groups happy. Perhaps there can be a look-up table to distinguish SP from TrEMBL?

  • Extended GO - we don't advertise this well. There's not even any documentation to say what's in this file and what makes it better than the existing file. What can we do to make this better? (Jane)

Do we want to make the extended files more prominent. Eventually the extended GO will become the main GO. We will push the consortium to see how close they are to being able to use the extended files.


Back to minutes list