search

geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
216 stars 39 forks source link

GO:0072703 cellular response to methyl methanesulfonate #16565

Open ValWood opened 5 years ago

ValWood commented 5 years ago

I'm wondering why we have this term:

GO:0072703 cellular response to methyl methanesulfonate

one of our community just used it, but this is really only a "condition" to invoke DNA damage. It should probably be "GO:0006974 cellular response to DNA damage stimulus" (and not used only for 'chemical sensitivities').....

Only 2 manual annotations IDA mouse IEP rat...

pgaudet commented 5 years ago

How about we merged it into "GO:0006974 cellular response to DNA damage stimulus"?

@slaulederkind @ukemi Does that work for you ?

ukemi commented 5 years ago

It depends on whether the response is to the chemical or the DNA damage that the chemical generates.

ValWood commented 5 years ago

But isn't "response to chemical" alone only a phenotype?, in this case you wouldn't make a GO annotation unless you could be specific about the process?

For example response to MMS in pombe, 344 genes and counting... https://www.pombase.org/term_single_allele_genotype_genes/FYPO:0000089

we wouldn't want to annotate these to a GO process, because we don't know what the underlying process is? usually it's just "cells die"

ukemi commented 5 years ago

Is it, what about pathways that transport toxins out of a cell or use a metabolic process to detoxify it? Is that a phenotype?

ValWood commented 5 years ago

So is the phenotype a transport defect? then it would be a process.

It just seems odd to curate "response to chemical" for every observation of chemical sensitivity, because they aren't saying anything about the process affected. ...But sure if you can be specific about the process (signalling, transport, detoxification).

ukemi commented 5 years ago

I have to admit, I've never been perfectly comfortable with the 'response' terms, but I do think that organisms have evolved processes that allow them to deal with chemicals in certain ways.

ValWood commented 5 years ago

I do think that organisms have evolved processes that allow them to deal with chemicals in certain ways.

I agree. But I think we should only be describing "response to chemical" if we can be a bit specific that this is an encoded process (and not just toxicity). To be able to differentiate between these 2 scenarios, more information is required. This should be enough information to annotate to some actual process describing the response.

We do this routinely:

x transport occurs_during response_to_x x translation occurs_during response_to_y x metabolism part_of response_to_y x detoxification part_of response_to_z etc. so we never need to use "response to" terms directly. This ensures we only use them when there is a real process involved.

I don't think a merge is appropriate, but I'm not sure that a specific term is required for a response unless a mechanism is identified.

It isn't very likely that there is a specific pathway for MMS, more likely a general pathway for allkylating agents? so it seems odd to curate a process for "response to every chemical/cancer drug" in this way.

RLovering commented 5 years ago

related ticket https://github.com/geneontology/go-ontology/issues/17227

ValWood commented 7 months ago

GO:0072703    cellular response to methyl methanesulfonate GO:0072703 cellular response to methyl methanesulfonate

Has ony 3 EXP, so I will proceed to obsoletion. These are just describing chemical sensitivities used in assays (i.e to induce DNA damage)

Fission yeast has 1100 genes https://www.pombase.org/results/from/id/1ee6e48c-5726-482d-85bf-9c5abedf01ad There is no GO term that you would want to annotate al of these to.

ValWood commented 1 week ago

@pgaudet obsolete or keep (close)?