grouping of pathogen and host morphology phenotypes

[ValWood]

Currently:

I think we need to avoid superclasses that group things like pathogen morphology and host morphology. Although these are related semantically by 'morphology' the are not really biologically related. the host morphology is different from the pathogen morphology.

Also,

label "host morphology phenotype during pathogen invasion definition "A pathogen host interaction phenotype which affects the host morphology during pathogen invasion. This relates to the pathogen colonization of the body, tissues, or cells of the host organism"

does not really belong under label "pathogen colonization of host phenotype" definition "A pathogen host interaction phenotype which affects the pathogen colonization of the body, tissues, or cells of the host organism."

becasue the latter is about the affects on pathogen colonization, not affects on the host morphology during the invasion process.

[CuzickA]

Hmm, yes tricky.

If I moved 'host morphology phenotype during pathogen invasion' branch out of "pathogen colonization of host phenotype" would it just go under 'pathogen host interaction phenotype' or an alternative parent? It looks like the current terms are at cell, tissue and organism level, so hard to know where to place it.

[ValWood]

It looks like the current terms are at cell, tissue and organism level, so hard to know where to place it. Yes this will be a problem, we should aim to get rid of these classifiers, they will make things too complicated.

I would make 'host morphology phenotype during pathogen invasion' a high level term for now, directly under 'pathogen host interaction phenotype'

Later maybe we will even have "host phenotype during pathogen invasion" That might be a more sensible way to group than tissue/cell etc.

[CuzickA]

Changed parentage of PHIPO:0001002 'host morphology phenotype during pathogen invasion' b2b5199

[CuzickA]

@ValWood should I obsolete the old grouping term 'pathogen colonization of host morphology phenotype' and have 'pathogen morphology phenotype during colonization of host' directly under 'pathogen colonization of host phenotype'?

[ValWood]

Yes that seems a good plan.

I find "'pathogen colonization of host morphology phenotype" quite hard to understand anyway! apart from the grouping of unconnected pathogen and host stuff.

[CuzickA]

Change PHIPO:0000998 pathogen morphology phenotype during colonization of host parentage 5f99f71

Obsolete PHIPO:0001007 pathogen colonization of host morphology phenotype a3c4b5a

Now looks like this

[CuzickA]

@ValWood what do you think about the 'abnormal host structure induced by pathogen' and 'host morphology phenotype during pathogen invasion'? Should these be under a shared parent term?

[ValWood]

what do you think about the 'abnormal host structure induced by pathogen' and 'host morphology phenotype during pathogen invasion'? Should these be under a shared parent term?

tricky. I don't think all structure development refers to morphology (size and shape). For instance the absence of seed development does not seem to be morphology. Also not all morphology phenotypes are abnormal "structures" (stunted size for example). I don't completely know how this should be modelled, but I suspect that some things will be both structure and morphology and these can have dual parentage (like membrane blebbing).

Another option is to have a "abnormal host development induced by pathogen" instead of
"abnormal host structure" induced by pathogen for the seed thing.

or to have "abnormal host seed development" modelled/classes as "abnormal biological process"

Let's wait for this part until, it might become more obvious the best way to do this as we add more subclasses, or a we can logically define these.

This could also be a question for Midori so I will make a note of this ticket.