search

geneontology / go-annotation

This repository hosts the tracker for issues pertaining to GO annotations.
BSD 3-Clause "New" or "Revised" License
35 stars 10 forks source link

PTN008708268 (PTHR46426) #4697

Closed sjm41 closed 1 year ago

sjm41 commented 1 year ago

The PROTEIN DISULFIDE-ISOMERASE TMX3 (PTHR46426) family comprises a single protein in all annotated species (except one), and it seems there's not a consensus on its function.

The node is currently annotated with the MF "protein disulfide isomerase activity' based on PomBase=SPBC3D6.13c (P87178) - this is an IMP annotation from PMID: 20204527.

But the human protein the family (TMX3/Q96JJ7) is annotated with protein-disulfide reductase activity - an IDA annotation from PMID: 15623505.

Maybe best not to have a MF on this node for the time-being? What do you think?

ValWood commented 1 year ago

I will take a look at the pombe annotation too. It shouldn't be MF by IMP unless there is additional supporting evidence.

marcfeuermann commented 1 year ago

Hi, Yes it seems wise to remove the PAINT annotation and wait for additional data. The IMP vs IDA question is a big question. From what I've seen so far, IDA is over-used. Many curators use IDA because they are convinced by the paper, regardless of the method used. I had endless discussions about this.

ValWood commented 1 year ago

I checked the paper and the annotation seems OK, it does appear to be a PDI. in vivo:

The fission yeast cells harboring pYPDI2 contained 1.62- and 2.73-fold higher PDI activity than the control yeast cells in exponential and stationary phases, respectively, indicating that the cloned gene is in vivo functioning.

Also, the intro states
"The oxidase activity is exhibited by PDI in the oxidized state, whereas the isomerase activity is displayed by PDI in the reduced state."

So presumably the activity of the PDIs is dependent on the oxidation state?

sjm41 commented 1 year ago

Thanks @marcfeuermann @ValWood Yes, I wasn't disputing the pombe annotation, just disputing whether it should be propagated to all orthologs given this single MF annotation and the 'competing' MF annotation on the human ortholog. Activity of this family could be species-specific or context-dependent (as you suggest) but it seems best not to propagate any MF via PAINT until we have more data.