-
Notifications
You must be signed in to change notification settings - Fork 2
dependent on - if we obsolete, what should we do with the current annotations? #17
Comments
General conclusions:
|
CC @rbalakri |
I think we need to be careful with capturing chemicals regardless of whether it is with dependent_on or in_presence_of. We need to consider if the chemical is physiologically relevant and not just an assay condition, for example using a chemical to induce a physiological response: if an author uses hydrogen peroxide to induce oxidative stress I would annotate to the relevant oxidative stress term rather than putting H2O2 in the extension. In the case of the ATP example above, we decided it shouldn't be added to the extension as ATP is actually mentioned in the reaction in the definition. In other cases we should consider creating new GO terms, e.g. ATP-dependent 'process/function'. Bottom line, the curator needs to start thinking a bit more about what physiological process is being demonstrated and not just blindly copy down what the author says. I'm still not sure we should keep this relation at all, I don't think I've yet come across a good example to justify keeping it. |
I agree with Rachael. Curators need to think about what the authors are saying. I would like to hold a workshop or tutorial at the GOC meeting (turns out the Symposium day is not packed after all) to highlight these issues. Rama |
Hi Rama An AE workshop would be brilliant, hopefully this would give us a chance to really work through examples Best Ruth From: rbalakri <[email protected]mailto:[email protected]> I agree with Rachael. Curators need to think about what the authors are saying. I would like to hold a workshop or tutorial at the GOC meeting (turns out the Symposium day is not packed after all) to highlight these issues. Rama Reply to this email directly or view it on GitHubhttps://github.com//issues/17#issuecomment-126031331. |
Hi All, So sorry I missed the call on Tue and because of our Dicty2015 meeting, I'll also miss Aug. 11. Thanks a lot for the notes, will go through when I get a chance. And yes, I’ll be in for an AE workshop on the symposium day! Best, On 30 Jul 2015, at 00:22, Ruth Lovering <[email protected]mailto:[email protected]> wrote: Hi Rama An AE workshop would be brilliant, hopefully this would give us a chance to really work through examples Best Ruth From: rbalakri <[email protected]mailto:[email protected]mailto:[email protected]> I agree with Rachael. Curators need to think about what the authors are saying. I would like to hold a workshop or tutorial at the GOC meeting (turns out the Symposium day is not packed after all) to highlight these issues. Rama Reply to this email directly or view it on GitHubhttps://github.com//issues/17#issuecomment-126031331. — |
Dependent_on
This has been used in 1800 annotations
It has been proposed that this relationship is depreciated
Suggestions need to be made to:
Examples
Many of these look like the extension information cannot be covered using the AE current relationships, and may be out of the scope of GO. Considerable guidance is required in order to ensure edits follow consistent approach, ie improved documentation
ChEBI ID
wide range of BP and MFs with ChEBI IDs. Possible activated_by or has_input?
Also note that this example is a substrate and that this is recorded in the ontology:
GO:0015616
DNA translocase activity
Catalysis of the reaction: ATP + H2O = ADP + phosphate, to drive movement along a single- or double-stranded DNA molecule.
GO ID
all BP and MF annotations C16 dependent_on GO IDs, consider changing to part_of, or reversing extension and using causally_upstream_of
would this one work in reverse, stating protein kinase activity part of necroptotic signaling pathway?
would this one work in reverse, stating antigen binding part of complement activation, lectin pathway?
all CC annotations C16 dependent_on GO IDs, consider changing to exists_during
No idea what to suggest with these
[RH] For the nucleosome example, I've looked at the paper and think there are better ways of capturing the data; see suggestions here http://wiki.geneontology.org/index.php/Annotation_Conf._Call,_July_28,_2015#i.29_PMID:_12482963
Pfam IDs
this hasn't been annotated to SH3 domain binding, so presumably this means that the protein annotated needs to have the SH3 domain to bind the target, is it necessary to state this? Is the the reciprocal annotation to SH3 binding
Protein IDs
is this suggesting that we list all proteins involved in a process in C16?
annotate the reverse ie O70362 protein localization to extracellular space C16 transports Q9ESD1
SO IDs
could this be Occurs_at?
[RH] Looking at the paper, Fos-1 and Hda-1 associate on the promoter of kreg-1 , so I think the correct way to represent this is with occurs_at, however you are not capturing the fact that it is the kreg-1 promoter. Is there a better ID that can be used to indicate this?
Converted from md report page created by @RLovering
The text was updated successfully, but these errors were encountered: