Ontology meeting 2015-10-01
Attendees:
Minutes: Paola
Annotation of 'mobile' gene products
Nature Plants article "Endogenous Arabidopsis messenger RNAs transported to distant tissues"
Produced chimeric plants by grafting shoots and roots of different A. thal ecotypes. The ecotypes differ by a large number of SNPS which can be used to discriminate the source of the transcript. RNA-seq was used on both root and shoot tissues in various graft combinations, found a total of 2006 mobile mRNAs. Transcript mobility was modulated by environmental stress. By detecting proteins with ecotype-specific seq in grafted tissue they identified proteins encoded by mobile mRNAs. But they also found non-self proteins that were NOT encoded by mobile mRNAs (meaning proteins had moved). --excerpted and paraphrased from accompanying News and Views article.
User's desire: wants to capture the 'mobile nature' of the mRNAs, be able to identify this subset of genes, preferably annotate with a GO cc term, original suggestion = "non-cell autonomous", evolved into "Cell-cell mobile protein/RNA"
TAIR recommendation: 'symplast' Question: can we add a related synonym/s of 'cell-cell mobile protein/mRNA/RNA/gene product' to 'symplast'? Question: what about a new term like 'cell-cell signaling via symplast-transported gene product'? Not for this particular data set but as a way to capture signaling that happens in this manner?
Cool! How about symplast-mediated signaling with two children symplast-mediated RNA signaling and symplast-mediated protein signaling? We also have RNA transport terms but in this case, what would be the agent? I don't think it would be the symplast because it seems to be the route the molecule takes.
DH: at MGI we’ve seen a similar case with mRNAs in neurons. We don’t do that, but we could have symplast-mediated signalling, with 2 children as suggested above, and use column 16 later.
PT: do they know that cells are present at all? TB: it is unlikely that cells are transported.
DOS: do we capture mobility at all, e.g. in blood?
DOS: does symplast even belong in GO? Shouldn’t it live in PO?
We could create a more granular term than symplast, e.g. differentiate the membrane part of the symplast from the rest. We could use more than one ontology. Use a new relationship/extension called ’translated_in’?
CM: add term in OBA and use that. It’s not really a job for GO, but we could help.
TB: what about the synonyms? DH: good, as related. DOS: they’re confusing. CM: that’s stretching too far just to make a text search work.
DH: use has_target_start/end_location?
Conclusion: TAIR can annotate to symplast, but can’t do much more. Could add gloss to the definition of symplast. Also, could use SO. SO has an attribute section. One is called ‘mobile’ and is defined quite generically. TB will look into that. No synonyms will be added to symplast.
Production Terms
It appears that these terms have been generating more discussion. Do we have a definite plan? Are we going to merge all the production terms into the regulation of production terms and change all those terms regulation of level terms? Is there a way to do this computationally rather than with manual merges?
DH leans towards making the ‘production’ terms do_not_annotate, and creating the regulation terms, and asking everyone to move their annotations there. Alex Diehl confirmed that’s what people want to capture, the regulation. Can we do that computationally? Yes, DOS could do it, and/or Heiko could create sort of a bulk TG job. DH will get an idea of how many terms we’re talking about. We could prepare a mapping table, though every group is likely going to be doing that differently.
Development Terms
Do we have a systematic plan to review the development/morphogenesis etc terms that do not have axioms?
AI: DH will look into this.
Side discussion: DOS find development overhaul very complex, easy to get lost. Should we review and come up with a simpler plan? Nobody has time. Add to list of future projects? Start off with use cases. We’d need a separate grant though.
Becky tested the 'morphogenesis' template. Her questions: 1) We’re now adding in ‘results_in_morphogenesis_of’ relationships (in addition to the cross-products)? 2) Many (but not all) other morphogenesis terms already in GO have ‘is_a: single-organism developmental process ; GO:0044767’ parentage. TG doesn’t currently suggest this parentage, and in most cases it would probably be inherited anyway.
David H tested all templates and suggests that we (Paola) can advertise them as soon as we've added the missing axioms (see above).
Adding logical defs to cellular response terms - bumped
DOS has started to do this. What's the syntax, and should we look into filling all gaps.
'cell proliferation' terms - bumped
Most have logical def, e.g. 'epithelial cell proliferation' is_a 'cell proliferation' that acts_on_population_of 'epithelial cell'. Did we have plans for a TG template? (Not in Jira.)
Protein fibril
Does it belong in GO? Stemming from a TG FF request where it is defined as "A polymer of proteins that form a fine fiber" and placed under 'protein complex'.
Bumped, but Harold confirmed that this was discussed with Alex Diehl previously. To the best of our memory, the point was that there is no stoichiometry for these things; this is what distinguishes them from being protein complexes. So they shouldn't be children of protein complex.