EBISPOT / efo

Github repo for the Experimental Factor Ontology (EFO)
https://www.ebi.ac.uk/efo/
56 stars 13 forks source link

EVA - EFO import #168

Closed hannah-mclaren closed 6 years ago

hannah-mclaren commented 6 years ago

Draft new terms for importing to EFO are in the 'EFOimport' tab of the following sheet: https://docs.google.com/spreadsheets/d/1wClVV3_wL_xJv6RB7yJYA5WlLvh0NzQGP_Pa4uEMUqY/edit?usp=sharing

Opening this issue ahead of submitting Webulous templates to get a second opinion on the parent terms. Most important are those highlighted in RED (these are high frequency >10 in clinVar). There are 9 terms i cannot find a parent for in EFO and most of these we did not import in the last submission either:

thyroxine-binding globulin quantitative trait locus blood group--lutheran inhibitor tyrosine kinase inhibitor response deoxygalactonojirimycin response prednisolone response warfarin response ivacaftor response - efficacy

I am away next week so aim to submit webulous templates after 4th June.

Other issues:

zoependlington commented 6 years ago

I have made a copy of your spreadsheet and completed some of the missing information which can be found here:

https://docs.google.com/spreadsheets/d/1ySfKKPvX-QR1GeKU8n0SAXrx3i4A9Ft3_92flA-Ek5Y/edit?usp=sharing

Any purple indicates changes to parent classification, bright yellow indicates any terms that did not have parent classification in the original spreadsheet and orange indicates terms that still do not have good classification. Pale yellow indicates that there are some notes written that sound be taken into account - typically in red. These are either notes to self or questions.

For the terms we didn't import last time:

I am still not sure about the classification of the terms with an asterix and require more research, right now the suggested parent terms are not ideal.

Edits to be made:

And finally, to answer your question:

I would create new terms under mental retardation (EFO:0003847) as they are different variants. They could also be added as subclasses of autosomal dominant non-syndromic intellectual disability (or autosomal recessive non-syndromic intellectual disability should there be any recessive types) alongside being subclasses of mental retardation.

This would reflect the DOID classification:

https://www.ebi.ac.uk/ols/ontologies/doid/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FDOID_0080231

hannah-mclaren commented 6 years ago

Webulous templates ready for review and import under: https://drive.google.com/drive/folders/1Nhlqhw2bpMPdEkyAJRk1sky7Gs4mIVl4?usp=sharing

Outstanding issues:

  1. Poor classification for the following terms. Do we import or exclude?
  1. Review EFO import phenotypes 05-06-2018

  2. Review EFO import disease 05-06-2018

  3. Review HP import 05-06-2018

  4. I found glycogen storage disease ii, adult form is a synonym of obsolete class [Orphanet:308604] which is replaced by EFO term Rare inborn errors of metabolism[Orphanet:68367]. Should we still go ahead with importing a new term for glycogen storage disease ii, adult form or should we exclude and EVA can map it appropriately on the next round of curation (since we have already submitted to OT)?

zoependlington commented 6 years ago

Thanks Hannah! I've reviewed and imported all the templates into EFO.

I've also imported thyroxine-binding globulin quantitative trait locus and blood group--lutheran inhibitor under the classes already agreed - the classification is correct just not overly specific. But I'll make a note to review these.

In EFO, Orphanet:308604 isn't obsolete and does have synonyms indicating GSD2 adult form, so I added "glycogen storage disease ii, adult form" as a synonym for EVA mapping and will make a note to review this when I return from holiday.