Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Obsolete terms #392

Open
jdferreira opened this issue Apr 2, 2019 · 3 comments
Open

Obsolete terms #392

jdferreira opened this issue Apr 2, 2019 · 3 comments

Comments

@jdferreira
Copy link

Some terms in this ontology are marked as obsolete, but not alternatives are given. For example, the majority of the interactions defined in BioGRID (https://thebiogrid.org/) are annotated the term MI:0799 ("additive genetic interaction defined by inequality"), which is obsolete.

Is there a recommended alternative for this term?

@pporrasebi
Copy link
Contributor

Dear user,

The PSI-MI ontology is a collaborative project where many experts in different areas contribute. We have contacted the person responsible for this specific set of terms to ensure the closest match to this term can be recommended.

For the time being, I would recommend to re-map it to either the generic 'genetic interaction' (MI:0208) or the more specific 'neutral genetic interaction' (MI:0934), both of which seem appropriate in this particualr case.

Please let us know if you have further questions, we will update this ticket as soon as we get feedback from the relevant experts.

@jdferreira
Copy link
Author

Thank you for the prompt reply and the suggestion for replacement.
I'll wait for the future reply.

@chris-grove
Copy link
Contributor

@jdferreira
I had worked with PSI-MI in the past to build the "genetic interaction" branch of the PSI-MI ontology. This ontology has undergone numerous revisions and there are some complicating (and contentious) factors in resolving the appropriate ontology structure.

I won't bore you with the details (unless you're really interested), suffice it to say that all (as far as I know or can tell) of the BioGRID genetic interactions labeled in their MITAB format download file as "additive genetic interaction defined by inequality" were originally curated to their own genetic interaction types "Phenotypic Enhancement" or "Negative Genetic", which indicates that a phenotype was enhanced (the organism had a more severe phenotype) in some way compared to a control, although we don't know if the enhancement was more than additive, additive or less than additive, etc. I believe the decision at BioGRID at the time (for that version of the genetic interaction ontology) was to map all "Phenotypic Enhancement" and "Negative Genetic" interactions to "additive genetic interaction defined by inequality". Unfortunately in the current version of the ontology there is no appropriate term to accurately capture the originally curated type as the definitions for "negative genetic" and "enhancement" are too strict given what is known from the original curation.

The remedy to this, which I'm working on, is to create terms that allow ambiguity as to whether the individual genetic perturbations combine additively or not (but aren't necessarily genetic interactions per se). I would recommend downloading (in addition to the MITAB file) the BioGRID TAB2 format file:

https://downloads.thebiogrid.org/Download/BioGRID/Release-Archive/BIOGRID-3.5.171/BIOGRID-ALL-3.5.171.tab2.zip

The specifications for which are here:
https://wiki.thebiogrid.org/doku.php/biogrid_tab_version_2.0

For all the "Negative Genetic" interactions in the Tab 2.0 file (see column 12), I think it's safe to say that these are all actually "negative genetic interaction":

https://www.ebi.ac.uk/ols/ontologies/mi/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FMI_0933

http://purl.obolibrary.org/obo/MI_0933

These "Negative Genetic" interactions make up the majority of the "additive genetic interaction defined by inequality" interactions from BioGRID (~544,000 out of 561,286) whereas the "Phenotypic Enhancement" interactions make up a minority (~17,200 out of 561,286). It is certainly possible that those remaining "Phenotypic Enhancement" interactions map to "enhancement":

https://www.ebi.ac.uk/ols/ontologies/mi/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FMI_1271

http://purl.obolibrary.org/obo/MI_1271

although, strictly speaking, they may not actually be genetic interactions (i.e. there was enhancement but it may simply have been additive and hence not a genetic interaction, strictly speaking), so buyer beware / caveat emptor.

I hope to soon have an updated genetic interaction ontology soon that should hopefully resolve some of these issues.

I hope that helps/clarifies things.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants