search

pombase / canto

The PomBase community curation tool
https://curation.pombase.org
Other
19 stars 7 forks source link

penetrance and expressivity #2399

Open ValWood opened 3 years ago

ValWood commented 3 years ago

We have "penetrance" as an extension option for both gene-for-gene and pathogen host-interaction phenotypes. We don't have "expressivity"

As most terms apply to whole plants, penetrance isn't usually applicable @jseager7 have we ever used it?

Is would seem that "expressivity" would be more useful (for example to qualitatively describe the extend of lesions).

Can anybody remember why we have it this way?

Val

jseager7 commented 3 years ago

Based on an export from 18 January, the has_penetrance extension has been used 21 times, and all of those are on PHIPO terms. I can get a list of sessions in which the annotations appear, but I think it would be better to open an issue on the PHI-base/curation if we want to make changes.

There should probably be an issue on this tracker or the curation tracker about why penetrance was enabled for our phenotypes in the first place. I can try and find that.

ValWood commented 3 years ago

They might be used in the single organism branch (we would want to keep it there)

I'm interested if they have been used for any pathogen host interaction phenotypes. it might be relevant there, I just can't think of an example.

jseager7 commented 3 years ago

Based on the latest export (27 January), it seems most of the phenotypes annotated with penetrance are pathogen-host interaction phenotypes:

It seems there's only one single species phenotype currently extended with penetrance:

ValWood commented 3 years ago

We need to review these. I suspect most should be "expressivity" not "penetrance".

CuzickA commented 3 years ago

I remember using penetrance in PHI annotations for bacteria killing nematodes, PHIPO:0001069 'death of host organism with pathogen' PHIPO:0001068 'survival of host organism with pathogen' some metagenotypes resulted in 100% death of population, whereas others resulted in 50% (or similar)

ValWood commented 3 years ago

OK that one is correct usage. We can probably limit penetrance to a few branches.

ValWood commented 3 years ago

JAmes could you let us know which sessions these terms have penetrance extensions: PHIPO:0000215 'host chlorosis present with pathogen' PHIPO:0001132 'increased size of pathogen vacuoles during host colonization' PHIPO:0000477 'presence of pathogen necrotrophic effector-mediated host programmed cell death' PHIPO:0001155 'decreased level of pathogen reactive oxygen species production within host'

These need to be checked to see if they should be expressivity. If so, penetrance can be restricted to these branches PHIPO:0001069 'death of host organism with pathogen' PHIPO:0001068 'survival of host organism with pathogen and others can be added if required.

Expressivity (severity) should be available for all of the abnormal terms in this branch (I think?)

ValWood commented 3 years ago

consider with https://github.com/pombase/canto/issues/2328

jseager7 commented 3 years ago

Here's the list of sessions with the above PHIPO terms, where at least one instance of the above terms has a has_penetrance extension:

ValWood commented 3 years ago

PMID:17041146

PHIPO:0001132 'increased size of pathogen vacuoles during host colonization' I removed this one. The text said 'hypheas occasionally had large vacuoles' (paraphrasing), but I don't really know whther this refers to one or more than one individual.  Penetrance would be OK here though.
PHIPO:0001155 'decreased level of pathogen reactive oxygen species production within host'

Penetrance is correct here. They looked at the proportion of the population.

So we probably need penetrance and expressivity(severity) in certain cases. We will just need to refine their avaiability as we go along, I think.