mRNA entities do not render in GO-CAM viz

[ValWood]

We have used some mRNA as inputs.

EG

mrh5C tranalationactivator complex has_input cox1.1

cbp7 translation activator has_input cob1.1

1. You can see the inputs in the menu, but not in the model.
2. It would be nice if these had an explicit "mRNA" label in case people do not understand the nomenclature

[ValWood]

model
http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel%3A67086be200000519

cc @kimrutherford

[kltm]

I can confirm that this is still true.
To clarify, this seems like has_part is generally not rendered.

[ValWood]

From discussions, this is a PomBase issue and a display issue.

1. We shouldn't be using transcript entities in this way.
2. The displays need to change so that it is somehow clear to the user whether an input entity is an mRNA or a protein. This is really critical when modelling pathways between transcription and translation. The models make no sense if we display a protein input before the protein has been created.

Once the display solution is in place, we'll change the entities back to protein. Otherwise a number of our current models in the mitochondrial project (and others) make no sense biologically.

I don't know if a display-level solution has yet been proposed.
@cmungall
@thomaspd
@vanaukenk
@hattrill

[cmungall]

I’m going to need a lot more context here Can we start by linking to the relevant parts of the GO-CAM docs for modeling transcription/translation?

…

On Fri, Oct 3, 2025 at 11:44 PM Val Wood ***@***.***> wrote: *ValWood* left a comment (geneontology/web-components#56) <#56> From discussions, this is a PomBase issue and a display issue. 1. We shouldn't be using transcript entities in this way. 2. The displays need to change so that it is somehow clear to the user whether an input entity is an mRNA or a protein. This is really critical when modelling pathways between transcription and translation. The models make no sense if we display a protein input before the protein has been created. Once the display solution is in place, we'll change the entities back to protein. Otherwise a number of our current models in the mitochondrial project (and others) make no sense biologically. I don't know if a display-level solution has yet been proposed. @cmungall <https://github.com/cmungall> @thomaspd <https://github.com/thomaspd> @vanaukenk <https://github.com/vanaukenk> @hattrill <https://github.com/hattrill> — Reply to this email directly, view it on GitHub <#56>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAMMONNJUVGVYBYXGAO4R33V5UC7AVCNFSM6AAAAAB2JWREROVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGNRXHE3DMMZSG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[ValWood]

ACTION: This is now a display issue, so we need to decide which terms display RNA input, and what to do for any activities which have both protein and RNA input (like adaptors).

[ValWood]

An example in this model
http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel%3A67086be200000519
Different translation regulators are specific for different mRNA inputs (indicated by cob1.1, atp9.1, cox1.1)