vesicular transport

[dosumis]

We have many transport terms that do not mention vesicles even though they involve/are mediated by vesicles. These terms have no classification that reflects this:

Source

ER - Golgi, PM -> endosome, endosome -> lysosome.

Shouldn't we fix this?

One confounding issue for patterns:

Do we refer to the vesicle as mediating the transport - or as the cargo? Essentially the same process transports the soluble contents of vesicles, the membranes (membrane trafficking) and proteins embedded in membranes (protein transport?). Can combine all these into general process terms - or do we keep separate hierarchies?

[dosumis]

Potential new design patterns:

'vesicle-mediated transport' and ('has target start location' some 'Golgi apparatus') and ('has target end location' some lysosome) and ('occurs in' some cytoplasm') # Strictly, need to say where goes in between. Could be buried in new genus { and ('transports or maintains some Y') # Optional to indicate cargo { and ('results in transport along' some Z } # Optional to indicate transport along cytoskeletal fiber

For membrane trafficking Y - membrane region For lipid trafficking Y = lipid (how to get link to membrane?) For protein trafficking Y = protein

Slightly radical proposal = merge all vesicle-mediated transport & vesicle transport terms. This may be hard to achieve. Maybe easier if we can come up with a relation that communicates that the vesicles are both a mediator of transport and something transported. Perhaps:

has_carrier': "A relationship between a transport or localization process and a continuant that mediates that process by physically carrying its cargo. More formally: X has_carrier Y iff: localization(X), continuant(Y), exists some cargo Z, continuent(Z), Z part_of Y, X has agent Y, X 'transports of maintains localization of' Y, X 'transports of maintains localization of' Z. SubPropertyOf: has_agent # - this needs to be merged with mediated_by SubPropertyOf: 'transports of maintains localization of'

@cmungall @ukemi @ValWood - Comments please.

[ValWood]

I'm not sure about this:

Slightly radical proposal = merge all vesicle-mediated transport & vesicle transport terms.

vesicle mediated-transport was intended for secretory pathway (membrane trafficking), (which I think uses actin?)

vesicle transport can also refer to GO:0047496 vesicle transport along microtubule

I'm not completely sure about the differences here. In pombe the tracking along a microtubule appears to be for proteins which are located to the cell tip on the internal side of the plasma membrane, polarity markers and such like (so not for things which are released, or located in the membrane) I don't know if this is a general distinction. Even though text books show the vesicles as cargo on microtubules, I have never seen any experimental evidence in fission yeast that the cargo transported along microtubules is actually in vesicles....although I've wondered about this. You might need a domain expert.....

Term Basket: 6

[dosumis]

Hi Val,

Don't see any issues with what you've posted. Clause to indicate transport (of vesicle) along cytoskeletal fibre is optional. The main idea is to collapse down the rather artificial distinction between transport of a vesicle and transport by a vesicle. Terms for transport directly along cytoskeletal fibers would not be affected

• David

[ValWood]

OK, I wasn't sure whether it was important.

[dosumis]

Pattern with new relation would be:

transport and ('has target start location' some fu) and ('has target end location' some bar) and ('occurs in' some CC') and (has_carrier some { vesicle or subclass of vesicle } ) and ('transports or maintains some CARGO ') and ('results in transport along' some { 'cytoskeletal fiber' or subclass } )

Still to be resolved:

1. Should the genus be transport or localization? Consistency is important for complete autoclassification.
   • Transport: "The directed movement of substances (such as macromolecules, small molecules, ions) into, out of or within a cell, or between cells, or within a multicellular organism by means of some agent such as a transporter or pore."
   • localization: "Any process in which a cell, a substance, or a cellular entity, such as a protein complex or organelle, is transported to or maintained in a specific location."
2. How to indicate when transport of a protein along a cytoskeletal fiber is direct - rather than via a vesicle? With the proposed pattern - the general class is agnostic about this. May need to be some manually maintained branch.

[dosumis]

Notes from editor's call:

"A distinction needs to be made on whether or not things are inside a vesicle and the vesicle is moving, or if things are moving in and out of the vesicle. [Does mediated by cover this?] ‘mediated by’ should mean that it’s not [sic?] necessary and sufficient for the process to happen, but as it is now, it’s probably not enough to distinguish between cases; definition is broad enough to cover both."

DOS: this seems like a red herring to me.

• mediated_by has no def. It should almost certainly be merged with 'has active participant' - which should probably be renamed (back) to 'has agent' (But note, we also have 'has causal agent'!). It doesn't really make sense to assert that a vesicle is the agent of transport in and of itself.

TBD:

CC @cmungall @tberardini @paolaroncaglia @ukemi @hdrabkin

1. Are there any objections to collapsing the distinction between vesicle-mediated transport & vesicle transport terms?
2. Are there any objections to defining the following new relation for use in the new, merged terms?
   • has_carrier: "A relationship between a transport or localization process and a continuant that mediates that process by physically carrying its cargo."
   • SubPropertyOf: has_agent # - this needs to be merged with mediated_by
   • SubPropertyOf: 'transports of maintains localization of'
   • More formally, X has_carrier Y iff:
   • localization(X),
   • continuant(Y),
   • exists some Z,
   • continuant(Z),
   • Z part_of Y,
   • X has agent Y,
   • X 'transports of maintains localization of' Y,
   • X 'transports of maintains localization of' Z.

\3. Should the genus be transport (which, according to the def, shouldn't really apply to vesicles, although it does right now) or 'establishment of localization'?

Note: a resolution soon would be very useful for synapse project: meeting next week.

[dosumis]

Chatted with @ValWood - she seems broadly in favour of collapsing down these distinctions. Also mentioned that membrane trafficking should be better supported - at least via synonyms.

[dosumis]

\3. Should the genus be transport (which, according to the def, shouldn't really apply to vesicles, although it does right now) or 'establishment of localization'?

transport: "The directed movement of substances (such as macromolecules, small molecules, ions) into, out of or within a cell, or between cells, or within a multicellular organism by means of some agent such as a transporter or pore." _isa: establishment of localization

establishment of localization: "The directed movement of a cell, substance or cellular entity, such as a protein complex or organelle, to a specific location."

Is there any way to broaden transport without it becoming indistinguishable from 'establishment of localization'? Not sure I can see one.

[ValWood]

Here are some terms which are 'establishment of localization' but which clearly involve no transport:

http://www.ebi.ac.uk/QuickGO/GTerm?id=GO:0070197#term=ancchart genes annotated to this are nuclear membrane and tether the telomeres.

http://www.ebi.ac.uk/QuickGO/GTerm?id=GO:0071790#term=ancchart or the spb

but this one sounds like it should be microtubule-mediated transport GO:0098863 nuclear migration by microtubule mediated pushing forces (is currently only establishment of localization)

[ukemi]

Establishment of localization can also include selective degradation/stabilization or tethering of a diffused substance.

[dosumis]

Establishment of localization can also include selective degradation/stabilization or tethering of a diffused substance.

Then this needs to be reflected in the definition, which not currently the case. First clause: "The _directed movement_ of ...."

[dosumis]

New suggested def

establishment of localization: "Any process that localizes a substance or cellular component. This may occur via movement, tethering or selective degredation."

[dosumis]

New suggested def

localization: "Any process in which a cell, a substance, or a cellular entity, such as a protein complex or organelle, is transported, tethered to or otherwise maintained in a specific location. In the case of substances, localization may also be achieved via selective degredation."

[ukemi]

This is better. We still need to figure out a way to distinguish between establishment and maintenance.

[dosumis]

No entirely happy with the selective degredation bits. @ukemi - do you know of any examples? IIRC, I tried to find some some examples in annotation when working on this branch last year, but failed to find any.

[dosumis]

This is better. We still need to figure out a way to distinguish between establishment and maintenance.

Maintenance certainly excludes movement. Tethering feels like it is both.

[dosumis]

I think this works as it is: maintenance of location: "Any process in which a cell, substance or cellular entity, such as a protein complex or organelle, is maintained in a location and prevented from moving elsewhere."

[ukemi]

That works for me.

[ukemi]

@tberardini Including Tanya here because many years ago we thought a lot about this. It was not simple.

[ukemi]

I think selective degradation/stabilization has been shown quite a bit in embryos. PMID:14588248 PMID:21035437 PMID:15186741

[dosumis]

It was not simple.

I think there are some fundamental problems to do with glossing of effects on concentration with location/localization of individual material entities. For example, transport of an entity changes the location of that entity, but may have not effect on the overall localization (local concentration) of entities of that type. Suspect we have to continue to gloss and work around these.

[tberardini]

Discussion getting v long. Can you consolidate the latest revised defs for the affected terms please, @dosumis?

[dosumis]

Consolidation of proposal so far:

Proposed new defs

(needed because current defs of localization and establishment of localization require movement as mechanism, should be broader than that):

establishment of localization:: OLD: "The directed movement of a cell, substance or cellular entity, such as a protein complex or organelle, to a specific location." -> NEW: "Any process that localizes a substance or cellular component. This may occur via movement, tethering or selective degradation." _isa: localization

localization: OLD: "Any process in which a cell, a substance, or a cellular entity, such as a protein complex or organelle, is transported to or maintained in a specific location." -> NEW: "Any process in which a cell, a substance, or a cellular entity, such as a protein complex or organelle, is transported, tethered to or otherwise maintained in a specific location. In the case of substances, localization may also be achieved via selective degradation."

Changing these will allow the definition of transport to be broadened to include transport of organelles (so vesicle transport is no longer a TPV):

transport: OLD: "The directed movement of substances (such as macromolecules, small molecules, ions) into, out of or within a cell, or between cells, or within a multicellular organism by means of some agent such as a transporter or pore." -> NEW: "The directed movement of substances (such as macromolecules, small molecules, ions) or cellular_components (such as complexes and organelles) into, out of or within a cell, or between cells, or within a multicellular organism by means of some agent such as a transporter, pore or motor protein." is_a: establishment of localization

Proposed new relation:

'has carrier': "A relationship between a transport or localization process and a continuant that mediates that process by physically carrying its cargo. More formally: X has_carrier Y iff: localization(X), continuant(Y), exists some cargo Z, continuent(Z), Z part_of Y, X has agent Y, X 'transports of maintains localization of' Y, X 'transports of maintains localization of' Z. SubPropertyOf: has_agent* SubPropertyOf: 'transports of maintains localization of'

*Currently called 'has active participant'. Needs to be merged with mediated_by

Resulting relation hierarchy:

• has_participant
  • has agent
  • has carrier
  • transports or maintains localization of
  • has carrier

This relation will allow us to merge vesicle-mediated transport branch vesicle transport branch

Proposed new pattern for transport

transport and ('has target start location' some fu) and ('has target end location' some bar) and ('transports or maintains some CARGO ') and ('occurs in' some CC') and ('has carrier' some { vesicle or subclass of vesicle } ) # optional clause and ('results in transport along' some { 'cytoskeletal fiber' or subclass } ) # optional clause

Potential problems:

• How do we distinguish direct transport along a 'cytoskeletal fiber', with a carrier? Leaving the 'has carrier' clause out just => a more general class.
• We have no class for vesicle-mediated protein transport. Should we add a branch for this? Would be useful for synapse project as they are very focussed on receptor transport/recycling via endosome. But retrofit to move relevant existing annotations down would be v.big.

[tberardini]

Suggest revisiting in GO eds meeting for consensus.

[paolaroncaglia]

Added to agenda.

[dosumis]

General issue for vesicle-mediated protein transport:

We currently have no term for this, although isolated specific examples. @valwood mentioned she always thought this would be useful. Currently annotates to 2process terms - a vesicle mediated one and a protein transport one. Not linked. Retrofit would be pretty huge.

Need to think about what parts vesicle mediated protein transport has that plain vesicle mediated transport does not.

[ValWood]

The history of this is that, as far as I am aware, at least for fission yeast every example of an annotation to "vesicle-mediated transport" annotation http://www.pombase.org/spombe/related/GO:0016192 could also be annotated to "intracellular protein transport".

However, even though all of these gene products are involved in protein transport some of the time, we cannot be sure that they always transport proteins. So certain instances of the annotations may not be correct. I am never sure if this matters if this is always a role of the gene product.....

Anyway, this is the reason that the vesicle mediated terms are not a descendent of intracellular protein transport. I usually make a concurrent annotation, but as you know I don't like making 2 annotations to describe the same process.

There are some 'protein specific' descendants like GO:0007038 endocytosed protein transport to vacuole which do have an intracellular protein transport' parent so currently the organization isn't very consistent.

[ValWood]

Wasn't Marc Feuermann adressing vesicel-meidated transport for PAINT (tagging @pgaudet as marc does not seem to have a GO git hub alias). He might have some insight on this topic....

[paolaroncaglia]

@marcfeuermann

[dosumis]

Changed defs as specified above, following agreement by GO editors

[pgaudet]

@dosumis @ValWood @paolaroncaglia @tberardini What remains to be done here?

[paolaroncaglia]

@pgaudet I think this is for @dosumis and @ValWood to address. Looks like @dosumis made some edits based on a consensus agreement, so perhaps this ticket can be closed? And a new one opened if there are (still) specific issues to address. Thanks.

[pgaudet]

Thanks @paolaroncaglia I see the edits, I'll close the ticket.