This is also mapping to membrane.
This location is very unlikely in yeast, as it is predomimnetly nuclear/nucleolar and other annotations implicate in histone modification. Has no tm domains. which organism does this come from?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
FK506 is a protein phosphatase inhibitor, and FK506 binding is a longstanding biochemical assay used to characterize certain types of proteins. The fact that FK506 is an immunosuppressant might be relevant if this were a BP annotation, but it is not relevant to this MF annotation. Also, there is at least one S. cerevisiae member of PTHR10516, FPR2, that is annotated with FK506 binding.
There should NOT be a taxon filter here.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Please note that "membrane" has many children that do not specify "integral to"; in fact, many of them specify "peripheral to" or something else indicating that the protein is close to but not in the membrane.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thats an odd one. I thought that FK106 was an immune system gene product when I read this but its a drug. Would we make this annotation in yeast? Binding of FK506 wouldn't be a normal function in yeast would it? Do we annotate binding to exogenous inhibitors (I don't know the answer but I haven't previously captured these with GO binding terms).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The physiological significance of FK506 binding in yeast may not be known, but clearly the protein can bind, so we left the annotation to the root. Please let us know if that's a problem.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
ignore previous post (deleted). FK506 is a drug isn't it.
which fission yeast gene were you annotating to this term?
I don't think we would capture this as a binding annotation if it is an inhibitor of a specific phosphatase), as it isn't related to the phosphatase activity in the wt. We usually capture inhibitors on the MF term in question with inhibited_by(CHEBI:xxxxx)
val
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I share your confusion about the whole FK506 binding issue. FK506 binds and inhibits yeast proteins like calcineurin and affects TOR signaling in yeast(s). It's not that the function of the yeast protein (like calcineurin) is to bind FK506, it's the other way around. FK506 binds the yeast proteins and significantly alters signaling pathways in several yeasts, including S. pombe. An inhibitor annotation does seem more appropriate.
Have you seen the fission yeast paper Ma et al, Genome-wide screening for genes associated with FK506 sensitivity in fission yeast. PLoS One 2011. A lot of genes/pathways seem to be modulated by FK506 in S. pombe.
Also note that Sc FPR2 is "Membrane-bound peptidyl-prolyl cis-trans isomerase (PPIase); binds to the drugs FK506 and rapamycin; expression pattern suggests possible involvement in ER protein trafficking; relocalizes from nucleus to vacuole upon DNA replication stress"
Sc FPR2 shows no evidence of a TM domain either even though it is membrane localized.
Diane
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Just to explain: we're annotating the FKBP family. According to http://en.wikipedia.org/wiki/Tacrolimus
"Tacrolimus reduces peptidyl-prolyl isomerase activity by binding to the immunophilin FKBP12 (FK506 binding protein)"- so technically FKBP binds (and presumably are inhibited) by FK506.
Unfortunately in PAINT we cannot make the 'extension' annotation 'inhibited_by', but this would be the best way to capture this. Meanwhile, I thought the 'binding' annotation was pretty harmless.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
PomBase wouldn't make this annotation because it isn't describing a function or process which occurs in wild type yeast. Its a 'condition' which produces signalling problems. I don't think its good practice to include this binding activity in GO annotation describing normal physiology. We might say that a specific enzyme could be inhibited by his compound, but anything else would be described using phenotype annotation.
Val
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I suspect the FK506 binding has only occurred because of the prominence of FK506 in the protein product name. There is a lot of literature on many protein-drug target interactions, but if we decided to include them it would change the scope of GO.
val
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Good point. However I propagated this because it was annotated across multiple species - and I think this is a common practice in the GO. We need to clearly define what the scope of GO is with respect to drug/small molecule binding. There are a lot of annotations affected. At the PAINT call agreed that Rama would take this to the annotation and ontology groups for wider advice.
Cheers,
Pascale
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Good point. However I propagated this because it was annotated across
multiple species - and I think this /is/ a common practice in the GO.
We need to clearly define what the scope of GO is with respect to
drug/small molecule binding. There are a lot of annotations affected.
At the PAINT call agreed that Rama would take this to the annotation
and ontology groups for wider advice.
Status: open Group: Labels: PANTHER annotation Created: Fri Apr 15, 2011 08:56 AM UTC by Valerie Wood Last Updated: Tue Jun 10, 2014 09:40 PM UTC Owner: Paul Thomas
is mapping to
FK506 binding
which is not appropriate for yeasts which don't have an immune sysyem.
There should be a taxon filter here if not already
I'm adding a link to this ticket within discussion to this ontology SF ticket: https://sourceforge.net/p/geneontology/ontology-requests/10930/
(mTOR-FKBP12-rapamycin complex)
and I'm marking that ontology ticket for discussion at our upcoming ontology SF jamboree.
Cheers,
Paola.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
At the last GO editors call and at the SF jamboree yesterday, we discussed the issue of having GO terms for complexes, binding and response to compounds that are exogenous to the species where the complex/binding/response is observed. Our conclusion is that we'd rather not allow this in the cellular component branch (e.g. 'mTOR-FKBP12-rapamycin complex', see https://sourceforge.net/p/geneontology/ontology-requests/10930/). However, exceptions would be made for host/symbiont complexes; these occur in natural situations, so are within scope of GO. We think that complexes with a drug bound are instead out of scope as that interaction would never occur naturally. The situation is a bit more complex for binding and response terms; we have many 'response to x' and 'x binding' terms where x is foreign to the organism and the interaction would also never occur naturally. However, in those cases we think it's relevant to look at the context where the binding/response are occurring. Therefore, as far as the ontology goes, we'd be ok with keeping the existing terms, including 'FK506 binding'.
We hope this makes sense. I'll send out an obsoletion email for 'mTOR-FKBP12-rapamycin complex'.
Thanks,
Paola
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This is also mapping to membrane.
This location is very unlikely in yeast, as it is predomimnetly nuclear/nucleolar and other annotations implicate in histone modification. Has no tm domains. which organism does this come from?
FK506 is a protein phosphatase inhibitor, and FK506 binding is a longstanding biochemical assay used to characterize certain types of proteins. The fact that FK506 is an immunosuppressant might be relevant if this were a BP annotation, but it is not relevant to this MF annotation. Also, there is at least one S. cerevisiae member of PTHR10516, FPR2, that is annotated with FK506 binding.
There should NOT be a taxon filter here.
As for membrane, the GAF for PTHR10516 shows that the annotation comes from 10 other proteins:
FB:FBgn0029174
MGI:MGI:95541
MGI:MGI:1336205
MGI:MGI:1341070
TAIR:locus:2089890
TAIR:locus:2100063
TAIR:locus:2135287
TAIR:locus:2181660
TAIR:locus:2194065
UniProtKB:P68106
Please note that "membrane" has many children that do not specify "integral to"; in fact, many of them specify "peripheral to" or something else indicating that the protein is close to but not in the membrane.
Thats an odd one. I thought that FK106 was an immune system gene product when I read this but its a drug. Would we make this annotation in yeast? Binding of FK506 wouldn't be a normal function in yeast would it? Do we annotate binding to exogenous inhibitors (I don't know the answer but I haven't previously captured these with GO binding terms).
Please refer to my earlier FK506 comment for answers to these question.
See https://sourceforge.net/p/geneontology/ontology-requests/10907/
Definition of the term GO:0005528 FK506 binding was changed to remove the reference to the immunosuppressing ability of FK506.
The physiological significance of FK506 binding in yeast may not be known, but clearly the protein can bind, so we left the annotation to the root. Please let us know if that's a problem.
ignore previous post (deleted). FK506 is a drug isn't it.
which fission yeast gene were you annotating to this term?
I don't think we would capture this as a binding annotation if it is an inhibitor of a specific phosphatase), as it isn't related to the phosphatase activity in the wt. We usually capture inhibitors on the MF term in question with inhibited_by(CHEBI:xxxxx)
val
Hi Val,
I share your confusion about the whole FK506 binding issue. FK506 binds and inhibits yeast proteins like calcineurin and affects TOR signaling in yeast(s). It's not that the function of the yeast protein (like calcineurin) is to bind FK506, it's the other way around. FK506 binds the yeast proteins and significantly alters signaling pathways in several yeasts, including S. pombe. An inhibitor annotation does seem more appropriate.
Have you seen the fission yeast paper Ma et al, Genome-wide screening for genes associated with FK506 sensitivity in fission yeast. PLoS One 2011. A lot of genes/pathways seem to be modulated by FK506 in S. pombe.
Also note that Sc FPR2 is "Membrane-bound peptidyl-prolyl cis-trans isomerase (PPIase); binds to the drugs FK506 and rapamycin; expression pattern suggests possible involvement in ER protein trafficking; relocalizes from nucleus to vacuole upon DNA replication stress"
Sc FPR2 shows no evidence of a TM domain either even though it is membrane localized.
Diane
Hello,
Just to explain: we're annotating the FKBP family. According to http://en.wikipedia.org/wiki/Tacrolimus
"Tacrolimus reduces peptidyl-prolyl isomerase activity by binding to the immunophilin FKBP12 (FK506 binding protein)"- so technically FKBP binds (and presumably are inhibited) by FK506.
Unfortunately in PAINT we cannot make the 'extension' annotation 'inhibited_by', but this would be the best way to capture this. Meanwhile, I thought the 'binding' annotation was pretty harmless.
I hadn't but I have upped the priority to process this one as we can now load phenotypic screen data fairly quickly.
thanks!
Hi Pascale,
PomBase wouldn't make this annotation because it isn't describing a function or process which occurs in wild type yeast. Its a 'condition' which produces signalling problems. I don't think its good practice to include this binding activity in GO annotation describing normal physiology. We might say that a specific enzyme could be inhibited by his compound, but anything else would be described using phenotype annotation.
Val
rapamycin binding is just as relevant for FK506 but nobody has annotated to that...
http://en.wikipedia.org/wiki/Mammalian_target_of_rapamycin
I suspect the FK506 binding has only occurred because of the prominence of FK506 in the protein product name. There is a lot of literature on many protein-drug target interactions, but if we decided to include them it would change the scope of GO.
val
Hi Val,
Good point. However I propagated this because it was annotated across multiple species - and I think this is a common practice in the GO. We need to clearly define what the scope of GO is with respect to drug/small molecule binding. There are a lot of annotations affected. At the PAINT call agreed that Rama would take this to the annotation and ontology groups for wider advice.
Cheers,
Pascale
OK will wait for the response
val
On 12/06/2014 16:34, Pascale Gaudet wrote:
Related
Annotation issues: #852
Hi,
I'm adding a link to this ticket within discussion to this ontology SF ticket:
https://sourceforge.net/p/geneontology/ontology-requests/10930/
(mTOR-FKBP12-rapamycin complex)
and I'm marking that ontology ticket for discussion at our upcoming ontology SF jamboree.
Cheers,
Paola.
Hi,
At the last GO editors call and at the SF jamboree yesterday, we discussed the issue of having GO terms for complexes, binding and response to compounds that are exogenous to the species where the complex/binding/response is observed. Our conclusion is that we'd rather not allow this in the cellular component branch (e.g. 'mTOR-FKBP12-rapamycin complex', see https://sourceforge.net/p/geneontology/ontology-requests/10930/). However, exceptions would be made for host/symbiont complexes; these occur in natural situations, so are within scope of GO. We think that complexes with a drug bound are instead out of scope as that interaction would never occur naturally. The situation is a bit more complex for binding and response terms; we have many 'response to x' and 'x binding' terms where x is foreign to the organism and the interaction would also never occur naturally. However, in those cases we think it's relevant to look at the context where the binding/response are occurring. Therefore, as far as the ontology goes, we'd be ok with keeping the existing terms, including 'FK506 binding'.
We hope this makes sense. I'll send out an obsoletion email for 'mTOR-FKBP12-rapamycin complex'.
Thanks,
Paola