WBConfCall 2014.04.03-Agenda and Minutes

From WormBaseWiki
Revision as of 16:30, 3 April 2014 by Cgrove (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search



Gene Association Files (GAFs)

  • Anatomy, life stage, phenotype, disease association files
  • Kevin Howe will be generating these files, in general
  • Will be difficult or impossible to abide by GAF rules exactly, but can work it out
  • Column 6 needs to be resolved: Reference or WormBase object (WBPaper or other) [primary source of the association]
  • Generally, each reference (paper) gets its own line in the GAF
  • Some anatomy terms are linked to multiple references, causes a problem
  • Phenotypes don't always have a reference, causes a problem
  • When we can't follow a publication, we reference a WB object
  • Sometimes GO annotations do not have a paper, and so use a different reference object, method
  • Proposal: Use a WormBase object as a reference (in column 6) for all GAFs aside from the GeneOntology GAFs
  • Kevin Howe: this can be problematic, not preferred
  • Disease terms can have multiple references and so each gets a separate line in the GAF
  • OMIM IDs can go in column 8; multiple OMIM IDs can be pipe separated
  • Anatomy, life stage, phenotype, disease GAFs: what is the referencing procedure? Currently using bogus WBPaper object (WBPaper00000000)
  • If we use a "Method"/"Analysis" approach, we could have a write up explaining where the data comes from
  • Anatomy GAFs can use ?Expr_pattern objects in column 6
  • Life stage GAFs can use RNA-Seq experiments in column 6
  • Disease GAFs will use ?Paper in column 6
  • Phenotype GAFs: use ?RNAi or ?Variation objects in column 6?
  • RNAi experiments all have WBPaper (or WBPerson at least), so that can be used; only ?Variation-to-?Gene connections are an issue
  • Many ?Variation-to-?Gene connections can be attributed to a WBPerson (e.g. from Allele curation form) in a Phenotype GAF
  • If a paper does not exist for ?Variation-to-?Gene or ?RNAi-to-?Gene connection, use the ?Variation or ?RNAi object in column 6 of the Phenotype GAF
  • Typically, a paper reference will go in column 6; WormBase data object (e.g. ?Variation or ?RNAi) will go in column 8
  • In the event that a paper reference does not exist, the WormBase object will go in BOTH column 8 and column 6