Tranporter Activity Meeting Notes
From GO Public
This is the header of the copy of the live file that I used when I took the branch:
date: 09:02:2007 17:43
auto-generated-by: OBO-Edit 1.002
subsetdef: goslim_generic "Generic GO slim"
subsetdef: goslim_goa "GOA and proteome slim"
subsetdef: goslim_plant "Plant GO slim"
subsetdef: goslim_yeast "Yeast GO slim"
subsetdef: gosubset_prok "Prokaryotic GO subset"
remark: cvs version: $Revision: 5.157 $
The development file is go/scratch/transport.obo.
The terms that were in the proposal carry the dbxref TR:jic so we can put on a global render and keep track of where they end up. I have pasted them directly into the live file so we can move them around and incorporate them with the live go file. New terms added during the meetings, including those in the initial proposal will carry the dbxref GOC:mtg_transport.
The number range in use is GO:0022801 to GO:0023000.
Implementation meeting: 19th February, 2007
Michelle Gwinn, Val Wood, Jennifer Clark. Minutes by Val Wood with additions by Jennifer Clark.
We started to work through the high level terms looking at which of the old terms corresponded to those in the proposal.
The top level substrate transporter terms are a bit tricky because they are defined as being broader than just transmembrane transporter terms so we are going to have to look at them on a case-by-case basis and figure out which have children that are not purely transmembrane transport (e.g. involve transport between cells or across multiple membranes). Those that have children like these will remain as children for transporter activity. Those that do not have children that are broader than simple transmembrane transporter activity terms will be moved to be under transmembrane transporter activity.
Following this we looked through some more top level terms and started to figure out which existing terms could be merged with the terms in the new proposal. This meant that we kept the good structure and correct definitions of the new terms but that, where possible, we kept the ids of the old terms. Several merges were made. The list can be generated by obodiff if required.
Jen to mail all terms which are just substrate to the annotation list to find out if they are used in any context other than simple transmembrane transport. - list extracted, not yet mailed.
Need to address 'chaperone' type activities (transport chaperone which 'binds and accompanies' also SAP receptors and 'SNARES' which are usually annotated as transporters
Next week need to address children of P-P bond hydrolysis driven transporter activity (it looks as though not all active transporters are carriers)
Propose the obsoletion of the remaining children of existing carrier activity
GO:0000036: acyl carrier activity Is undefined but is used for annotating ACPs so is not a transmembrane type carrier. We later decided just to move this up under transporter activity and think more about it later. [DONE]
GO:0017077: oxidative phosphorylation uncoupler activity
Undefined, I think ther reason was that we decided this wasn't a carrier, btu I can't remember exactly
Note: After discussion on the GO list this term was properly defined both in the live GO and in the draft file. This term does not need to be obsoleted. [ACTION ITEM DONE]
GO:0008320: protein carrier activity This has been used for karyopherins and other proteins which are not transmembrane carriers (the term carrier appears to have multiple useage). The karyopherin type of protein transporter is indistinguishable for protein carrier. Maybe this should be a merge into protein transporter rather than an obsoletion.....
The next meeting will be 2pm GMT, 9am EST on Friday 23rd February.
Implementation meeting: 23rd February, 2007
Michelle Gwinn, Val Wood, Jennifer Clark, Ian Paulsen.
Minutes by Val Wood.
Merged carrier GO:0005386 with proposed term 'primary active transporter'.
Merged uniporter GO:0015292 with proposed term facilitated diffusion.
Destroyed proposed term ‘passive carrier’.
Merged 'permease activity' with ‘transmembrane transporter activity/GOnew’.
Merged group translocator with protein-N(PI)-phosphohistidine-sugar phosphotransferase activity GO:0008982.
Created new specific children of ion transporter activity under 'active transport'.
Renamed efflux permease activity to ‘efflux transmembrane transporter activity’.
Renamed ‘uptake permease activity’ to ‘uptake transmembrane transporter activity’.
For each child of : protein-N(PI)-phosphohistidine-sugar phosphotransferase activity (GO:0008982) [X] permease activity OR [X] porter activity
i) merge into parent term xxx transporter activity
(Note that after the e-mail to check that nobody meant these to refer to any form of non-membrane transport, that these parent terms will become ‘xxx transmembrane transporter activity’)
ii) remove parentage to protein-N(PI)-phosphohistidine-sugar phosphotransferase activity
iii) Create a new child term of BOTH the GO:0008982 and the previous merged term with standard term name and definition of the type protein-N(PI)-phosphohistidine-[X] phosphotransferase activity
iv) add PTS transporter synonyms
2. Standardize definitions and term names of all children of efflux and uptake terms to ‘efflux transmembrane transporter activity’ and to ‘uptake transmembrane transporter activity’ to remove ‘permease' from term names.
3. Action item Deal with all term names containing ‘permease’ ‘carrier’ or ‘porter’ as above
Add permease and carrier porter synonyms globally where appropriate
The next meeting will be 2pm GMT, 9am EST on Friday 26th February.
Implementation meeting: 26th February, 2007
i) GO:0015583 beta-glucoside [arbutin-salicin-cellobiose] permease activity
Reason: groups 3 substrates
We created new terms for each substrate, the current term has no annotations.
[AGREED AND DONE]
ii) GO:0015266 protein channel activity [NOT DONE]
and its child GO:0015266 apoptogenic cytochrome c release channel activity [DONE]
Reason: These are undefined and we don't know what they mean.
iii) GO:0015557 arginine targeting transporter activity
Reason: We don't know what this is.
Later comment by Val: looking at this the arginine targeting appears to refer to arginine residues in the leader motif of the translocated proteins.
It probably refers to the The TatABCE system. It translocates several redox enzymes to the /E. coli/ periplasm
not sure what we want to do with this
iv) MAPK phosphatase transporter activity and MAPK transporter activity
Reason: Term refers to specific gene producs but we don't know which, or what type of transporter. No annotations. [DONE]
v) oxidative phosphorylation uncoupler activity
Reason: Term is undefined. Doesn't make sense as a transporter.
[REVERSED] This term was subsequently defined in both the live and draft files.
Other Action items
1) Action item: Mail list with general query about transporter terms containing 'drug' (Michelle)
Can they be merged with toxin? or some other way to show in the tem name that these refer to endogenous/exogenous substances.
GO:0015238 drug transporter activity
GO:0015307 drug:hydrogen antiporter activity
2) Add reasons for obsoletion to existing proposal on Wiki (Val)
3) may need to fix 'ATPase activity, coupled to movement of substances'
Should they all be transmembrane?
4) DNA translocase activity, need definition GO:0015616 which mentions transmembrane movement?
5) SF 1666307 (Val and Jen)
need to work out how the mitochondrial 'translocases' relate to
GO:0015450: protein translocase activity
GO:0015462: protein-transporting ATPase activity
GO:0008566: mitochondrial protein-transporting ATPase activity
GO:0016464: chloroplast protein-transporting ATPase activity
This is related to SF 1666307 so we need to check that we have a transporter function term appropirate for these mitochondrial 'translocases'
(Which should be under the new macromolecute transporter activity?)
(Val: it isn't clear which are P-P bond hydrolysis, may beed to separate out some of these?)
6) Ask Candace to propose defs for type I-> type IV transporters (Michelle).
The next meeting will be 2.30pm GMT, 9.30am EST on Monday 5th March.
Implementation meeting: 19th March, 2007
Present: Jennifer Clark, Val Wood, Ian Paulsen, Michelle Gwinn.
Merged new 'channel activity' with existing 'channel or pore class transporter activity' 0015267
Started looking at all children of channel
We fixed all the children of alpha-type channel activity and fixed all the relationships of the immediate children under channel activity.
Moved porin activity to be a child of 'wide-pore gated channel' term
Moved gap junction activity to be a child also.
Looked at children of porin
Suggested obsoletions from above work
i) receptor porin (combines 2 functions/ is a gene product term)
ii) voltage-gated anion channel porin activity /JUST WRONG :)
iii) alpha-type channel / defined based on a general structural distinction (alpha -helix containing) rather than a functional distinction
iv) monocarboxylic acid transporter activity 0008028 and all children [DONE]
P.S. I think this was actually monocarboxylate channel activity ; GO:0015256 and children. These are the terms that have been obsoleted and it was double checked. The above was an error in the minutes.
v) NADPH oxidase-associated cytochrome b558 hydrogen channel activity (gene product)
vii) protein channel activity
The gene products annotated to this term are not 'protein channels' /we are not sure what a protein channel is/the term is undefined
viii) apoptogenic cytochrome c release channel activity (This seems to be a gene product.)
i) remove 'carrier' from all defs and replace with active transporter. This is really an action item to apply aristotelian defs to all children of active carrier
have done all of these that have been looked at and found to be okay. [DONE]
ii) deal with 'ionophore' terms. There was some confusion about how to define these and Michelle or Ian were going to look them up.
- mobile ion carrier activity: this sounds like a 'chaperone' of some sort...
in fact I'm not sure that it isn't related to the term 'metallochaperone activity' 0016530 (Maria Costanzo might be able to point us in the right direction here)
iii) Check all remaining children of channel
iv) Check all children of porin
We had terrible sound interference on skype but this was solved by reinitiating the call from Jennifer's computer instead of Val's and by getting Val to use her internal mic and mute whilst not talking.
The next meeting will be 2.30pm GMT, 9.30am EST on Monday 26th March.
Implementation meeting: 26th March, 2007
Transmembrane Electron Carrier Activity
'transmembrane electron carrier activity'
Query: Is this related to:
electron carrier activity
Any molecular entity that serves as an electron acceptor and electron donor in an electron transport system
Note: there are also lots of obsoleted electron carrier related terms
Why isn't nuclear transport through the nuclear pore considered transport across a membrane?...I think this is because the substrate does not travel *through* the membrane (i.e the membrane is completely excluded from the nuclear pore (it would be considered nuclear transport if it went from *through* the inner or the outer membrane? see:
Does everyone agree with this distinction, and is it clear why it is excluded from transmembrane transporter activity from the current def?
Jen: I think it's because transport across the nuclear membrane is a process with several steps whereas the usual transmembrane transporter activity terms are just transfering things in one step across one membrane.
We looked at the chaperone terms and decided that it would be best to work on these as a separate progect after transport. Please see the chaperone wiki page for details.
Started to work through all terms containing 'permease'
Generally merge with the equivalent 'transporter' term which is usually a parent. e.g. copper permease would be merged with the parent copper transporter activity.
(Note some people use permease to refer to 2ndary active transport. This is not general and not the way the terms are currently defined
merged amino-acid-polyamine transporter into amine transporter
ACTION ITEM JEN
amino acid transport tidy/standardize defs to reflect parentage (i.e adding transmembrane?)
ACTION ITEM JEN
Make all L-amino acid permease/transport synonyms and primary names just <amino acid> without the 'L-'
ACTION ITEM JEN
Create BOTH an ATPase and a 2ndary active transporter parentage for all amino acids (we need to discuss how to do this more next week.
ACTION ITEM Jen
Colour all things we have addressed a different colour for next week so we can see how much there is to go and prioritize a bit more clearly.
Continue working through permeases
Implementation meeting: 2nd April, 2007
action item: Ask Ian about 'narrow pore defintion'.
action item: move ATPase activity, coupled to movement of substances from under "transmembrane transporter activity"
(note from Val, I question the need for this term, because when queried previously its direct annotations were spurious, we need to check into this, |Jen can you note that this needs looking into?)
action item: Query functional difference between
"ATPase activity, coupled to transmembrane movement of ions GO:0042625"
"ATPase activity, coupled to transmembrane movement of ions, phosphorylative mechanism GO:0015662"
We think we know the distinction (i.e. no 2 is for F1/F0 ATPase but we don't think this is so clear from the defs
finish working through and standardising all terms and defs for transmitter gated ion channel and parents and children
Between-Meeting Progress - 4th April
I have just moved amino acid transporter activity under transmembrane transporter activity. I have added 'transmembrane' to all the descendant names and applied the standard definition. I left the high-affinity terms. I have not marked any of these terms as complete with the dbxref as they have not been examined in detail by the experts yet.
Between-Meeting Progress - 11th April
I have pulled out the list of terms that remain undefined and circulated them to the list so we can all mark terms that obviously need obsoleted.
I have also sent this proposal:
The transporters are currently grouped under:
drug transporter activity
neurotransmitter transporter activity
odorant transporter activity
toxin transporter activity
vitamin transporter activity
xenobiotic transporter activity
x substance transporter activity
(e.g. amine transporter activity)
It seems to me that is would be best if they are just grouped under substance specificity and that any of the other groupings are just handled by giving a synonym to the child terms. Would you be happy with that?
For example, instead of:
[i]neurotransmitter transporter activity
---[i]acetylcholine transporter activity
we could have:
[i]acetylcholine transporter activity
broad_synonym: neurotransmitter transporter activity
(Example spotted by Alex Diehl.)
This would seem preferable, because although many species may be transporting acetylcholine they may not all be using it as a neurotransmitter.
I am not sure about these two:
cofactor transporter activity
siderophore transporter activity
That do you think about this idea? Do you think that some would be okay to change and others not? I know that xenobiotic, toxin and drug have been difficult to change in the past but we could just do some and leave the more difficult ones.
Jclark 07:13, 11 April 2007 (PDT)
Between-Meeting Progress - 12th April
I am working through fixing the permease terms. Many of the are defined in text form and by their position in the graph as symporters and so I am moving the permease name to be a synonym and just changing the name to show that they are symporters. In cases where the term is undefined and has no dbxref or parentage showing anything other than the substrate of the transporter I am just merging the term with the relevant transporter or transmembrane transporter activity.
I have realised at this point that following this work we will need to remap the general dbxrefs to the TC database as the work we are doing will have shown up errors in this mapping.
Implementation meeting: 16th April, 2007
We worked through the list of undefined terms and worked a plan for how to deal with most of them.
Between-Meeting Progress - 17th April
I have feedback on how to deal with these terms:
type I protein secretor activity type II protein secretor activity type III protein (virulence-related) secretor activity type IV protein (DNA-protein) secretor activity type V protein secretor activity
Candace Collmer forwarded your question regarding definitions for the various bacterial secretion activities. Over the past year I have worked on development and refinement of biological process and cellular component terms for the various bacterial secretion systems and the question of what to do with molecular function has been discussed at some length.
As someone who has spent most of my research career working with various bacterial secretion systems, it is my opinion that "secretor activity" molecular function terms are not appropriate for secretion systems I-IV. Each of these secretion systems is composed of a complex of proteins or set of protein complexes which together accomplish various aspects of the complex assembly and/or stages in the secretion function, but with no single protein having autonomous "secretor activity". Instead, individual components of these systems can be annotated to a cellular component ( the secretion complex/complexes) and biological process (protein secretion by the type # pathway) - annotations which I believe fully capture the concept that "secretor activity" attempts to encompass. In select cases where more is know about the specific biochemical/enzymatic functions of individual components, annotation to other molecular functions may be appropriate. For example, GspE in the Type II pathway could be annotated to the ATP binding (GO:0005524.)
The inappropriateness of "secretor activity" molecular function terms is further highlighted by the fact that this phrase is not used at all by researchers in the bacterial secretion field and is therefore unlikely to be used.
In contrast to secretion systems I-IV, the type V system could more plausibly be annotated to a molecular function "secretor activity" as secretion is accomplished by a domain of the protein being secreted.
The biological process definition for "protein secretion by the type V secretion system" is as follows:
The process by which proteins mediate their own secretion across the outer membrane through a beta-barrel pore structure formed by the C-terminal domain of the protein precursor. Following passage across the outer membrane, the mature protein is released from the pore by an autocatalytic activity. Proteins secreted by the Type V system are first translocated across the plasma membrane by the Sec pathway. [source: GOC:pamgo_curators]
Perhaps a molecular function term for the type V pathway could read:
Formation of a beta-barrel pore structure by the C-terminal domain of the protein precursor that mediates secretion of the mature protein across the outer membrane
But again, I don't see this as adding significantly to the information captured by the biological process term. In short, I guess I'm ultimately recommending that Molecular Function terms specifying "secretor activity" be obsoleted for all bacterial secretion systems.
let me know your thoughts on this,
Magdalen Lindeberg ml16 at cornell.edu
On 5th June the obsoletion mail was sent with the following explanation.
N.B. Obsoleting only the function terms; cellular component terms for the same systems will be left in place.
Function ontology term obsoletion reasons: - The community does not use the terminology employed - Use of an "umbrella" molecular function term is ill-suited to several of the secretion systems.
Consequence: - A number of the proteins annotated to the cellular component terms for the secretion complexes will have unknown molecular function - This accurately reflects the state of the field, and specific molecular functions (binding, cleaving, etc) can be added as they are characterized.
Implementation meeting: 30th April, 2007
I had a skype call with Peter D'Eustachio and he says that he'd be happy for all the substrate specific term under transporter activity to move under transmembrane transporter activity.
Implementation meeting: 5th June, 2007
1, Move SNAP receptor and children from under intracellular transporter to under protein binding (done)
2. Add to SF request to define SNAP terms (Val)
3. Propose obsoletion of secretor complex I -> V function terms (see forwarded e-mail) (Jen)
This is the summary of the rasons behind it:
"Thanks for your thoughtful response. In light of your comments, I support leaving the cellular component terms in place and not keeping the molecular function terms (for the reasons you suggest - that the community does not use the terminology employed, and that use of an "umbrella" molecular function term is ill-suited to several of the secretion systems). I have no problem with the idea that a number of the proteins annotated to the cellular component terms for the secretion complexes will have unknown molecular function - in fact, I think that accurately reflects the state of the field, and that specific molecular functions (binding, cleaving, etc) can be added as they are characterized."
4 GO:0015638 microcin uptake permease activity . Ask Ian if he knows what this is to define.Jen is not sure whether it should be a symporter (Jen)
5. Make sure symporter and ABC transporters for both directions standardized to IN or OUT and OR or AND. Use defs of carbohydrate transporting ATPase as template. (Jen)
Node standards in docs (Jen)
6. Propose obsoletion of 'GO:0015096 manganese resistance permease' Unused term. Don't know mechanism so can't put in best position in existing graph. Best to request a new term when used (Jen)
7. Check standard defs. http://wiki.geneontology.org/index.php/Docs (Val and Michelle)
8. Terms to mail transport interest group and ask for suggested defs for ( we don't know what they are, outside our collective area of expertise): (Jen)
dihydropyridine-sensitive calcium channel activity GO:0015270 (used by human and mouse)
store-operated calcium channel activity GO:0015279 (human/rat/worm/fly/mouse)
volume sensitive anion channel GO:0005225 (used by Drosopila)
A bunch of terms containing the word 'rectifier' (used by human, mouse, rat, drosophila):
ATP-activated inward rectifier potassium channel activity GO:0015272
delayed rectifier potassium channel activity GO:0005251
G-protein activated inward rectifier potassium channel activity GO:0015467
G-protein enhanced inward rectifier potassium channel activity GO:0015273
inward rectifier channel (Obsolete) GO:0005241
inward rectifier potassium channel activity GO:0005242
open rectifier potassium channel activity GO:0005252
outward rectifier potassium channel activity GO:0015271
excitatory extracellular ligand-gated ion channel activity GO:0005231
Note: Although there is only one direct annotation (Drosophila ISS) this term has
children used for many annotations (mouse/rat/zfish/worm) which DO NOT appear
to specify 'excitatory' ....should they?
amiloride-sensitive sodium channel activity GO:0015280
(used by mouse/rat/zebrafish/fly/human)
9. mail transporter interest group and give 2 weeks notice that the following substrate specific terms will move under transmembrane transporter activity unless anybody shouts loudly: (Jen)
nucleobase, nucleoside, nucleotide and nucleic acid transporter activity - Done
organic acid transporter activity - Done
carbohydrate transporter activity - Done (still a few descendants that need more extensive overhaul. Search on 'enable'.)
alcohol transporter activity - Done
10. move auxin (efflux and influx ) under as these must be transmembrane (Jen) - done
11. mail Tanya about chlorophyll catabolite (Jen) chlorophyll catabolite transporter activity GO:0010290
Implementation meeting: 11th June, 2007
Michelle Gwinn, Jennifer Deegan
Term: low affinity zinc ion tranporter activity This terms says in the def 'powered by proton motive force'. We are not sure what this means. Does it mean that hydrogen ions pass through the membrane in the other direction or does it just mean that the membrane potential of the imbalance in hydrogen ions forces the zinc across. Where does the 'probably' come in?
The standard defs that we have in the docs are ready to be implemented.
the non-directional transporters can be standardised to say (in/out).
Bug fixing meeting: 31st Oct, 2007
Participants: Doug Howe, Jennifer Deegan.
We discussed an error that Doug has noticed in one of our edits and the following email was subsequently sent to the annotation list:
During the transport work we made, and then fixed, a small incorrect edit that altered some annotations. As a consequence, we think you might possibly want to adjust your annotations to make them more granular. If you do not make the adjustment, the annotations will still be correct, they will just be less granular.
In more detail, during the transport work we erroneously merged a child term into its parent, and subsequently made a new child term that exactly replaced the term that had been subsumed in the merge. The annotations that were moved as a consequence of the merge are not now wrong, but they are no longer as granular as they were before. Specifically, the old term 'connexon channel activity' ; GO:0015285 has been subsumed into gap junction channel activity ; GO:0005243, and then connexon channel activity ; GO:0015285 was replaced with the new term 'gap junction hemi-channel activity' ; GO:0055077 which is a part_of child of 'gap junction channel activity' ; GO:0005243. If you would like to take the annotations that were once to 'connexon channel activity' ; GO:0015285 and move them to 'gap junction hemi-channel activity '; GO:0055077 then that would put them back to their more granular state. If you leave them on 'gap junction channel activity'; GO:0005243 they will still be correct but will be less granular than they once were.
I hope we have not caused too much inconvenience with this edit. Please write back if you have any questions.