cytokine activity

[gocentral]

I'm wondering what "cytokine activity" means as a molecular function. Essentially, a cytokine binds its receptor, and then the transduction signal takes over in the realm of biological process. Is "cytokine activity" a biochemical/biophysical molecular function? I question the validity of this term and "growth factor activity" and "hormone activity" in the molecular function aspect. They sound more like biological process terms.

Reported by: slaulederkind

Original Ticket: geneontology/ontology-requests/6269

[gocentral]

• labels: --> Term obsoletion

Original comment by: slaulederkind

[gocentral]

I'm inclined to agree, but I've asked Alex Diehl to comment because he's well versed in this area.

m

Original comment by: mah11

[gocentral]

Stan,

These terms have been discussed on a number of occasions and always been kept in the ontology because of their evident utility and continuing ontological correctness. The argument is that the 'activity' is an intrinsic feature of certain gene products that is easily assayed, just as an enzymatic activity is an intrinsic feature of particular gene products that is easily assayed. The basic assay for a cytokine is to throw a molecule onto to some cells in culture or inject in vivo, look for various changes in cellular or organismal processes, such as differentiation or proliferation, and upon finding changes, state that the molecule has activity as a cytokine. In ontological terms, we would say the function of the molecule is to act as a cytokine, which the GO has named 'cytokine activity'. The function of a cytokine is to induce biological processes, but is not a biological process in of itself. Thus it does make ontological sense for these terms to be part of the molecular function ontology. This interpretation is consistent with the definition of what a function is in relation to biology, as described by Arp and Smith (http://precedings.nature.com/documents/1941/version/1) (There is a longer version of the paper by Arp and Smith being submitted now, which Barry can provide upon request -- I'm reluctant to give it out prior to publication).

I don't deny that GO interpretation of what the molecular function ontology represents is veering off towards representing molecular processes in some cases, but many GO MF terms can still be interpreted as true functions in the sense of Arp and Smith. Probably the fact that most MF terms can be interpreted as having the ability to carry out simple enzymatic or binding events is the source of the confusion here, but molecular function needs to be interpreted more broadly than that in order to be consistent with the BFO sense of biological function. The GO needs to remain consistent with BFO.

In practical terms, there are significant numbers of annotations to these terms, and it would be a major loss of information to obsolete the terms and delete the annotations. The terms provide a useful way of finding gene products that can act as cytokines, growth factors, and hormones, and often appear in microarrays analyses that enrich for GO terms.

Thanks,

Alex

Original comment by: addiehl

[gocentral]

Alex -

Thanks for your thorough comments.

I'll admit to leaning more towards the camp that would prefer to bring the MF ontology into consistency with a scope of processes in the BFO sense. But that decision is far from final, so I'm willing to keep the cytokine, hormone, and growth factor terms in MF for the present.

m

Original comment by: mah11

[gocentral]

closing, because even if there is any change, it will be a long way off.

Original comment by: mah11

[gocentral]

• status: open --> closed-works-for-me

Original comment by: mah11

[gocentral]

• status: closed-works-for-me --> closed-later

Original comment by: mah11