Ontology meeting 2016-07-21
Attendees: Melanie, David OS, David H, Peter d'E, Chris, Tanya
- switched to https://build.berkeleybop.org/job/release-go-v2/
New release pipeline is in place. go plus should not get ahead of rest of files.
- how is the inference report working out for people?
Good. If blank, no new inferences but has script has run.
- OE save filter issues
Not a problem anymore. Was a timing issue. Good to use again now. Pls. also use P5 :)/
- P5 memory settings - all good?
Not as stable as P4 but usable.
interPro wants to release new version of slim, but would like to maintain access to old one. Should version IRI PURLs resolve? Which relations should be kept? E.g. http://www.geneontology.org/ontology/subsets/goslim_metagenomics.obo
Melanie redid GOslim that INTERPRO is using - changed the ontology itself by updating tags. Changes should filter through to the online file (URL above).
IRIs: Chris: looks like it uses the old style IRIs. To point to the old file, need to use a different IRI. (Melanie/David - please add details.) Chris: Who are the users? MC: Interpro Metagenomics users! https://www.ebi.ac.uk/metagenomics/projects/SRP033553/samples/SRS512695/runs/SRR1045093/results/versions/3.0 Click on Functional Analysis tab.
Relations: No has_part. At most basic, is_a and part_of. Chris will take care of this GH ticket
is_a: GO:0003674 ! molecular_function
relationship: has_part GO:0003676 ! nucleic acid binding
relationship: part_of GO:0008150 ! biological_process
Assertion or inference?
Should we assert the is_a or let the reasoner do the job? Also confirm Jenkins check is disabled.
If logical def exists, then no asserted is_a needed.
name: RNA localization to nucleus
def: "A macromolecular localization process in which RNA is transported to and maintained in a location within the nucleus." [GOD:mah, PMID:26305931]
synonym: "RNA localisation to nucleus" EXACT [GOC:mah]
is_a: GO:0006403 ! RNA localization
intersection_of: GO:0006403 ! RNA localization
intersection_of: has_target_end_location GO:0005634 ! nucleus
Request for obsoletion
Val said "I often wonder with terms like this, if there is no difference between the process DNA replication initiation and DNA replication initiation from late origin except via regulation
should the term DNA replication initiation from late origin exist, or should we only have the regulation terms in GO?"
Midori said "The mechanism and involved gene products are the same at early origins and at late origins, so there is no differentium to distinguish "initiation at late origin" from "initiation"."
Why obsolete and not just merge? Request 'late origin' from SO so that we can use that to refer to for the regulation terms. See GH ticket for additional details. GH12533
Xrefs to Reactome
Looks like this wasn't fully addressed yet. See email thread here: While Peter is on site at Jax, we should devise and action plan for this.
https://github.com/geneontology/go-site/issues/198 File another ticket on the GO tracker that updates the main GO file, assign to Chris. All species xrefs or only human? Human is backed by evidence. Other species likely electronic inference from human annotation. Use human only, start there.
And this GH issue:
Review capitalisation of terms
Not an emergency, push to next meeting. https://github.com/cmungall/capital-offence