geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
220 stars 40 forks source link

Define starts and ends for integrated stress response signaling (ISR) #29042

Open cmungall opened 6 days ago

cmungall commented 6 days ago

currently there are 36 genes in human annotated to this which seems high

The definitions are inconsistent in how they are phrased, and whether they talk about starts or ends:

I propose we model ISR as starting with the activity of the member of the eIF2a kinase family.

Note this would rule out annotations to e.g DELE1 https://amigo.geneontology.org/amigo/reference/PMID:32132706

However, an argument could be made for including this (the def for the PKR form starts "starts with the activation of..."). But we should be clear and consistent!

The end point is trickier. The PERK form is the only one to talk about the end and it says a generic ends with regulation of a downstream cellular process, e.g. transcription. In this case, this could be anything. It could be argued that the endpoint is the transcription regulated by ATF-4 but some might consider it to be extended to include GADD34.

Tangentially related: https://github.com/geneontology/noctua-alliance-pathway-preview/issues/16

ValWood commented 5 days ago

@PCarme we can put integrated stress response on our target pathway list since fission yeast is used as a complementary model our PI works on this ! https://pubmed.ncbi.nlm.nih.gov/33313903/

Also tagging @sylvainpoux who has done a lot of work on the human annotation for this area.

So we have 2 experts.