inconsistent relationships between flower and inflorescence

[edenny]

Hi @ramonawalls and @ehartley ,

In reviewing the traits.csv file that Emily created after the last release, I noticed that some of the inflorescence classes have a link with flower classes and some don't.

If {open inflorescence} is present then {open flower} is present
If {pollen-releasing inflorescence} is present then {pollen-releasing flower} is present

However,

If {inflorescence}, {non-senesced inflorescence}, {unopened inflorescence}, or {senesced inflorescence} is present, there is nothing that says there is a {flower}, {non-senesced flower}, {unopened flower}, or {senesced flower} present.

Should we include that relationship for any of those traits?

[ramonawalls]

That relationship should be inferred, so we may need to add an equivalency axiom. Ramona

…

------------------------------------------------------ Ramona L. Walls, Ph.D. Executive Director of Data Science, Critical Path Institute Assistant Research Professor, Bio5 Institute, University of Arizona Part-time artist (https://www.instagram.com/omnivoresgratitutude/) *I respectfully acknowledge that my home and offices are on the land andterritories of Indigenous peoples. Today, Arizona is home to 22 federallyrecognized tribes, with Tucson being home to the O’odham and the Yaqui.*

On Mon, Jun 23, 2025 at 12:14 PM edenny ***@***.***> wrote: *edenny* created an issue (PlantPhenoOntology/ppo#140) <#140> Hi @ramonawalls <https://github.com/ramonawalls> and @ehartley <https://github.com/ehartley> , In reviewing the traits.csv file that Emily created after the last release, I noticed that some of the inflorescence classes have a link with flower classes and some don't. If {open inflorescence} is present then {open flower} is present If {pollen-releasing inflorescence} is present then {pollen-releasing flower} is present However, If {inflorescence}, {non-senesced inflorescence}, {unopened inflorescence}, or {senesced inflorescence} is present, there is nothing that says there is a {flower}, {non-senesced flower}, {unopened flower}, or {senesced flower} present. Should we include that relationship for any of those traits? — Reply to this email directly, view it on GitHub <#140>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAZIHLS5QYDASLDUC4PGTRT3FBGZBAVCNFSM6AAAAAB76HPV3GVHI2DSMVQWIX3LMV43ASLTON2WKOZTGE3DSMRQGI2TCNI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[edenny]

Similarly some of the {reproductive shoot system} classes are linked with the corresponding flower class and some are not so we should make that consistent too.

I also noticed in Protege that while {non-senesced reproductive shoot system present} shows up as a child to {reproductive shoot system present}, {senesced reproductive shoot system present} does not show up as a child there. I only see it under {senesced flower present} but I think it should be in both places.

[elhartley]

Similarly some of the {reproductive shoot system} classes are linked with the corresponding flower class and some are not so we should make that consistent too.

I also noticed in Protege that while {non-senesced reproductive shoot system present} shows up as a child to {reproductive shoot system present}, {senesced reproductive shoot system present} does not show up as a child there. I only see it under {senesced flower present} but I think it should be in both places.

I think the links to the corresponding flower classes should be removed from the reproductive shoot system classes. For instance, it is incorrect to say a senesced reproductive shoot system is a subclass of has visible part some senesced flower since flowers aren't the only type of reproductive shoot system. Instead, I think the flower, inflorescence, and strobilus classes should be made subclasses of the corresponding reproductive shoot system class (e.g., make senesced reproductive shoot system the parent of senesced flower, senesced inflorescence, and senesced microsporangiate strobilus).

[edenny]

Logically what Emily says makes sense. But this gets a little tricky now that we added gymnosperms, because originally this group was {floral structure} and was intended to be used where counts could have been either of angiosperm flowers or inflorescences (which we have in the NPN dataset), but it was assumed to include that basic unit of a flower.

Since a lot of users are going to be looking at flowering phenology and the {flower present} trait, I think we want to make sure {flower present} is somehow linked to a trait that encompasses either flower or inflorescence counts.

I am not sure the best way to do this. Perhaps include another layer of hierarchy below {reproductive shoot system} that separates gymnosperms and angiosperms...?

Any thoughts @ramonawalls ?