obophenotype / human-phenotype-ontology

Ontology for the description of human clinical features
http://obophenotype.github.io/human-phenotype-ontology/
Other
289 stars 51 forks source link

abnormal bone ossification is not morphology but should be physiology #9761

Closed LCCarmody closed 5 months ago

LCCarmody commented 1 year ago

HPO term HP_0011849 Abnormal bone ossification

Suggested revision and reasons Abnormal bone ossification eq uses GO process term and describes the abnormal process of ossification. We propose moving it to be a child of 'Abnormal musculoskeletal physiology'. All ossification child terms would move with to, except 'Abnormality of bone mineral density.

Suggestion for new structure:

Abnormality of the skeletal system |   |   |   |   |   -- | -- | -- | -- | -- | --   | Abnormal musculoskeletal physiology |   |   |   |     |   | Abnormal bone ossification |   |   |     |   |   | Abnormal enchondral ossification |   |     |   |   | Abnormal intramembranous ossification |   |     |   |   | Decreased skull ossification |   |     |   |   | Ectopic ossification |   |     | Abnormal skeletal morphology |   |   |   |     |   | Abnormal bone structure |   |   |     |   |   | Abnormality of bone mineral density |   |   Also, suggestion to revise EQ statement from 'has part' some (quality and ('characteristic of' some (ossification and ('occurs in' some 'bone element'))) and ('has modifier' some abnormal)) To 'has part' some (quality and ('characteristic of' some ossification) and ('has modifier' some abnormal)) BECAUSE 'Ectopic ossification' is a child term and ectopic ossification occurs in a part that is not a bone element. If you are okay with this, I will make the changes. Working with @sbello @rays22 et al., looking to harmonize HP and MP.
pnrobinson commented 1 year ago

Clinically these are items that are almost always inferred from a single observation, which we refer to be the result of a process but we are not actually measuring a process and I think this placement is not what clinicians would expect. You can measure the time course (process) of ossification experimentally pretty easily but this is not what the HPO is modeling.

LCCarmody commented 1 year ago

Then how about we re-write the logical definition so that it is not a process but in terms of morphology. And re-write the term as the result. 'Abnormal increased bone deposition'? also, can we get rid of 'bone element' in the logical def, as that is not true in the 'ectopic ossification'.

pnrobinson commented 1 year ago

I think the best thing would be to the PATO term for ossified https://www.ebi.ac.uk/ols/ontologies/pato/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FPATO_0001448

sbello commented 10 months ago

@pnrobinson @LCCarmody Do you also plan to revise the text definition to make it clear you are defining the outcome of the process rather than the process itself. I'm going back through the closeMatch mappings and came across these again. I'm planning to change the mappings to relatedMatch since the MP term is really about the process and if I understand correctly the HPO term is about the outcome of the process. I think it would be helpful to have that distinction in the text definitions as well as in the logical ones. Thanks, Sue

pnrobinson commented 10 months ago

In clinical medicine we are not observing the process but we are observing the outcome. I am guessing this is also true for at least some uses of MP terms. Our rule is essentially that processes are things that are measured over time (movies) rather than in one instant (pictures). I agree it would be good to go over the definitions, but maybe we can try to do this is a systematic way with a workshop or some colleagues in skeletal medicine.

pnrobinson commented 5 months ago

OK, I think we can close this...please reopen if needed.