PAMGO ML modifications

From GO Wiki
Jump to: navigation, search

Contents

Biological Process

Proposed term modifications

(only items to be changed are shown)

type I protein secretion system ; GO:0030253

Synonyms

  • broader: ABC translocator
Marcus

It appears to me that the extant terms for protein secretion systems (GO:0030253, GO:0015628, GO:0030254, GO:0030255) are named inappropriately for the biological process ontology. As worded, a "type I protein secretion system" is a thing (a complex of gene products); it is not a process or activity, e.g. attribute of the thing. Many obsolete GO terms were made obsolete because they represent gene products. For example, consider the following:

  • GO:0008014 calcium-dependent cell adhesion molecule activity
  • Definition: OBSOLETE. A calcium-dependent cell adhesion protein (type I membrane protein) that interacts in a homophilic manner in cell-cell interactions.
  • Comment: This term was made obsolete because it represents a gene product. To update annotations, use the biological process term 'calcium-dependent cell-cell adhesion ; GO:0016339', the cellular component term 'integral to membrane ; GO:0016021' and the molecular function term 'protein binding ; GO:0005515'.

Appending the word “activity” did not rescue GO:0008014 from obsolescing, I assume, because the definition described the protein rather than the process. I see strong parallels in the case of the type I-V protein secretion systems, for example:

  • GO:0030253 type I protein secretion system
  • Definition: A bacterial secretion system in which a complex of three proteins spans the inner membrane, periplasmic space, and outer membrane; does not involve proteolytic processing of secreted proteins.

The definition above describes the thing that does the secreting, rather than the process of secretion. Therefore, I think these protein secretion system terms (GO:0030253, GO:0015628, GO:0030254, GO:0030255) should be made obsolete, and new terms created that reflect the process rather than the thing. As an example, I suggest one of the following as a replacement for type I protein secretion system:

  • type I protein secretion system-dependent secretion (no… kind of bulky)
  • secretion via type I protein secretion system (no… sounds like dual annotating with a process and a component)
  • type I protein secretion (yes… simplicity is my favorite)

As a definition for type I protein secretion (the replacement for GO:0030253) I might suggest some minor re-wording of the obsoleted definition:

  • Definition: Secretion of protein via a complex of three proteins spanning the inner membrane, periplasmic space, and outer membrane of a cell, which does not involve proteolytic processing of the proteins secreted.

(Note: I am under the assumption that we even need this term because we want to be able to annotate the protein being secreted, for example an “effector,” not just the protein components of the secretory complex. Otherwise why not just dual annotate to GO:0009306 protein secretion and one of the appropriate protein secretion system complex terms, e.g. GO:0030256?)

I cut “bacterial secretion system” from the definition because I believe (I could be wrong—anyone care to correct me?) that whenever possible we should not specify organisms in definitions. For type V protein secretion system, would it be necessary to say “Gram-negative bacteria” or could that be cut?

I also suggest adding a comment to enhance clarity, such as the one for GO:0005234 glutamate-gated ion channel activity, which is not obsolete:

  • GO:0005234 glutamate-gated ion channel activity
  • Comment: Note that this term represents an activity and not a gene product. Consider also annotating to the molecular function term 'glutamate receptor activity ; GO:0008066'.
Jane

Hi Marcus - all these changes to the secretion system terms are fine, but do we need to obsolete the existing terms? Would it be okay if we just changed the names and definitions?

Magdalen

Jane, I'm not sure if my response to Marcus's comments that I emailed around 9-15-06 and to which Trudy responded here on the wiki have been incorporated into your modifications, so at the risk of repetition, I'll paste the biological process terms for protein secretion developed in response to Marcus's comments below. Also, in reponse to your question, I'm fine with changing the names and definitions and not obsoleting the terms. Note that my comments in this round are in blue.

Protein secretion by the type I secretion system

Accession: GO:0030253

Ontology: biological_process

Synonyms: exact: Protein secretion by the TOSS

Definition: The process by which proteins are secreted into the extracellular milieu via the type I secretion system; secretion occurs in a continuous process without the distinct presence of periplasmic intermediates; does not involve proteolytic processing of secreted proteins; (Bacteria, ncbi_taxonomy_id:2).

Comment: Note that this term represents an activity and not a cellular structure. Consider also annotating to the cellular component term ‘type I protein secretion system complex; GO:0030256'.


Protein secretion by the type II secretion system

Accession: GO:0015628

Ontology: biological_process

Synonyms:

exact: protein secretion by the T2SS

exact: protein secretion by the T2S

other-related: protein secretion by the Sec-dependent system [sensu bacteria]

other related: protein secretion by the general secretion pathway [sensu bacteria]

other related: protein secretion by the general secretory pathway [sensu bacteria]

Definition:

The process by which proteins are secreted across the outer membrane by the type II secretion system. Proteins using this pathway are first translocated across the plasma membrane via the Sec (general secretory) or Tat pathways. (Bacteria, ncbi_taxonomy_id:2).

Comment: Note that this term represents an activity and not a cellular structure. Consider also annotating to the cellular component term ‘type II protein secretion system associated complexes; GO:0015627'.


Protein secretion by the type III protein secretion system

Accession: GO:0030254

Ontology: biological_process

Synonyms: exact: protein secretion by the TTSS

exact: protein secretion by the T3SS

exact: protein secretion by the T3S

Definition:

The process by which proteins are transferred into the extracellular milieu or directly into host cells by the type III secretion system; secretion occurs in a continuous process without the distinct presence of periplasmic intermediates; does not involve proteolytic processing of secreted proteins (Bacteria, ncbi_taxonomy_id:2).

Comment: Note that this term represents an activity and not a cellular structure. Consider also annotating to the cellular component term ‘type III protein secretion system complex; GO:0030257'.


Protein secretion by the type IV protein secretion system

Accession: GO:0030255

Ontology: biological_process

Synonyms: exact: protein secretion by the T4SS

Definition:

The process by which nucleoprotein DNA conjugation intermediates or proteins are transferred into the extracellular milieu or directly into host cells, via the type IV protein secretion system.

Comment: Note that this term represents an activity and not a cellular structure. Consider also annotating to the cellular component term ‘type IV protein secretion system complex; GO:_____’

Protein secretion by the type V protein secretion system

Accession: GO:0046819

Ontology: biological_process

Synonyms: exact: protein secretion by the autotransporter system

Definition:

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

Comment: None

Jane

Thanks Magdalen - I've implemented all of those changes now. I removed the reference to the bacterial taxon from the first three defs because we only usually include this in GO where the term has a 'sensu' ending - if you want me to restore the 'bacterial' bit from the original definitions, let me know. The other thing was that the term names varied between using 'protein secretion system' and just 'secretion system'. I presumed this wasn't deliberate, and standardised them all to say 'xxx secretion system', each with 'xxx protein secretion system as an exact synonym. I this isn't correct, please let me know.

Magdalen

Jane, your solution described above looks great. The original terms in GO used the phrase "protein secretion system" in the primary term names, a phrase which is not used in the literature. Consequently, these terms were hard to find when searching.

type II protein secretion system ; GO:0015628

Synonyms

  • exact: type II protein (Sec) secretion system
  • exact: T2SS
  • exact: main terminal branch
  • exact: MTB
  • related: Sec-dependent secretion system [sensu bacteria]
  • related: general secretion pathway [sensu bacteria]
  • related: general secretory pathway [sensu bacteria]

Definition:

  • A bacterial system for secretion of proteins across the outer membrane using a dedicated secretion apparatus involving multiple proteins. Proteins using this pathway are first translocated across the inner membrane via the Sec (general secretory) or TAT pathways.
Marcus

See comments above regarding thing vs. process

type III protein secretion system ; GO:0030254

Synonyms

  • exact: TTSS
  • exact: T3SS


Marcus

See comments above regarding thing vs. process

type IV protein secretion system ; GO:0030255

Synonyms

  • exact: T4SS


Marcus

See comments above regarding thing vs. process

Type IV pilus dependent-motility ; GO:0043107

(primary term name was switched with that of a synonym, since I think its better to not use acronyms (aka TFA) as the primary term name)

Synonyms:

  • exact: TFP-dependent motility
Jane

Done.

pilus biogenesis ; GO:0009297

I have tweaked the definition of the following term so that individual pili types (such as Type IV pili) can be children. As the definition was previously written, it implied that all children had to be involved in conjugation, adherence and so on…

Synonyms:

  • exact: pilus biosynthesis

Definition:

  • The assembly of a pilus, a short filamentous structure on a bacterial cell, flagella-like in structure and generally present in many copies. Pili are variously involved in transfer of nucleic acids, adherence to surfaces, and formation of pellicles Is required for bacterial conjugation and plays a role in adherence to surfaces (when it is called a fimbrium), and in the formation of pellicles.
Jane

Done.

Proposed new terms

Sec protein translocation system

  • Child of “Intracellular protein transport” GO:0006886
  • Child of “Protein insertion into membrane” GO:0051205

Synonyms:

  • exact: Sec protein export system
  • exact: Sec-dependent protein translocation
  • related: general secretion pathway
  • related: general secretory pathway
  • related: general export pathway

Definition:

  • A system for translocation of unfolded proteins across membranes, composed of a conserved heterotrimeric complex together with various accessory proteins. As in, but not restricted to, the taxon Bacteria (Bacteria, ncbi_taxonomy_id:2). In bacteria, Sec-translocated proteins are subsequently secreted via the type II, IV, or V secretion systems.
Marcus

I think that the proposed term sec protein translocation system needs to be reworded to reflect the process of which it is a part (see 1.1.1.1 above). Currently it sounds like a “thing” or gene product, rather than a process.

Jane

Any suggestions as to what we should call it then? Sec protein-mediated translocation? Protein translocation via Sec translocation system?

Magdalen

Again, reinterating from my response to Marcus of last summer, I propose:

Protein transport by the Sec protein translocation system (sensu Bacteria)

Accession:

Ontology: biological process

Synonyms: exact: protein transport by the Sec protein export system other related: protein transport by the general secretion pathway [sensu Bacteria] other related: protein transport by the general secretory pathway [sensu Bacteria] other related: protein transport by the general export pathway [sensu Bacteria]

Definition:

The process by which unfolded proteins are inserted into or translocated across the plasma membrane by the Sec protein translocation system. As in, but not restricted to, the taxon Bacteria (Bacteria, ncbi_taxonomy_id:2).

Comment: Note that this term represents an activity and not a cellular structure. Consider also annotating to the cellular component term ‘Sec complex; GO:0031522'.


Twin-arginine translocation system

  • Child of “Intracellular protein transport” GO:0006886

Synonyms

  • exact: "Tat protein translocation system"
  • exact: "twin-arginine-dependent translocation"

Definition:

  • A system for translocation of folded proteins across membranes, generally involving three integral membrane proteins. As in, but not restricted to, the taxon Bacteria (Bacteria, ncbi_taxonomy_id:2). In bacteria, proteins translocated by the Tat pathway may be subsequently secreted via the type II secretion systems.
Marcus

I think that the proposed term twin-arginine translocation system needs to be reworded to reflect the process of which it is a part (see 1.1.1.1 above). Currently it sounds like a “thing” or gene product, rather than a process.

Jane

Again, any suggestions as to what we should call it?

Magdalen

Protein secretion by the Tat protein translocation system (sensu Bacteria)

Accession:

Ontology: biological process

Synonyms: none

Definition:

The process by which folded proteins are translocated across the cytoplasmic membrane by the Tat protein translocation system. As in, but not restricted to, the taxon Bacteria (Bacteria, ncbi_taxonomy_id:2). Proteins translocated by the Tat pathway may be subsequently secreted via the type II secretion system.

Comment: Note that this term represents an activity and not a cellular structure. Consider also annotating to the cellular component term ‘TAT protein translocation system complex (sensu Bacteria); GO:_____’


Type IV pilus biogenesis

  • Child of “pilus biogenesis”: GO:0009297

Synonyms:

  • exact: Type IV fimbria assembly
  • exact: Type IV fimbria biogenesis
  • exact: Type IV fimbriae assembly
  • exact: Type IV fimbriae biogenesis
  • exact: Type IV fimbrial assembly
  • exact: Type IV fimbrial biogenesis
  • exact: Type IV fimbrium assembly
  • exact: Type IV fimbrium biogenesis
  • exact: Type IV pilus biosynthesis
  • exact: TFP biogenesis
  • exact: Type 4 pilus biogenesis

Definition:

  • The assembly of a Type IV pilus, composed of a pilus fiber and approximately ten proteins at its base; Type IV pili play a role in cell motility, adherence to substrates, and aggregation
Jane

Done. GO id = GO:0043683

Cellular Component

Proposed term modifications

(only items to be changed are shown)

type I protein secretion system complex ; GO:0030256

Synonyms:

  • broader: ABC translocator complex
  • Change from being child of GO:0044459 “plasma membrane part” to child of GO:0044464 “cell part” (The decision to move large secretion complexes directly to “cell part” GO:0044464 was made with Michelle prior to the July meeting)
Jane

Done.

Marcus

A question regarding type I protein secretion system complex:

  • GO:0030256 type I protein secretion system complex
  • Definition: A complex of three secretory proteins that carry out secretion in the type I secretion system: an inner membrane transport ATPase (termed ABC protein for ATP-binding cassette), which provides the energy for protein secretion; an outer membrane protein, which is exported via the sec pathway; and a membrane fusion protein, which is anchored in the inner membrane and spans the periplasmic space.

Is the ATPase considered a “secretory protein”? If not, I suggest tightening up the definition. Is the ATPase membrane-bound? Does it have to be, or just the complex of which it is a part? (I agree with moving its parent term from plasma membrane part to “cell part.”)

Jane

Anyone have any further comments on this? I have left the def as is for the moment.

Magdalen

Secretion systems I-IV are each composed of multiple proteins. Do all members of each complex qualify as “secretory proteins”? Well, if one defines “secretory protein” as a component of a complex that, when absent, prevents the complex from carrying out secretion (my favored definition), then yes they are. But if “secretory protein” implies physical interaction with the proteins being secreted, then only a subset of proteins in these complexes would be considered secretory proteins. I consider the inner membrane ABC protein to be a secretory protein because it is a dedicated component of a secretion system complex. (That is, it doesn’t float around providing energy for other processes). It also happens to be membrane associated, although given that the three proteins form a complex spanning the inner and outer membranes, the precise location of this single component, or any other, should not affect assignment of the Type I secretion system complex term to the parent “cell part” rather than “plasma membrane” or “cell envelope”.


type II protein secretion system associated complexes ; GO:0015627

  • Change from being child of GO:0044459 “plasma membrane part” to child of GO:0044464 “cell part”

Synonyms:

  • exact: type II protein (Sec) secretion system associated complexes
  • exact: T2SS associated complexes
  • exact: main terminal branch
  • exact: MTB
  • related: Sec-dependent secretion system associated complexes [sensu bacteria]
  • related: general secretion pathway associated complexes [sensu bacteria]
  • related: general secretory pathway associated complexes [sensu bacteria]

Definition:

  • A set of complexes composed of two or more of the approximately 15 proteins that together carry out Type II protein secretion across the outer membrane, following transport across the inner membrane by the Sec or TAT pathways
Jane

I'm a bit confused by this - GO classes are always singular, so for example if I wanted a term for all RNA-containing complexes, I would call it 'RNA-containing complex' (not RNA-containing complexes) and its definition would be 'A protein complex includes a molecule of RNA' (not 'Protein complexes that contain RNA'). So with that in mind, would it be okay to change the term above to:

  • name: type II protein secretion system associated complex
  • definition: A complex composed of two or more of the approximately 15 proteins that together carry out type II protein secretion across the outer membrane, following transport across the inner membrane by the Sec or TAT pathways.

? I've changed the parentage as suggested though.

Magdalen

I think your proposed change in the definition will work.

Request to move mis-annotated genes out of GO:0015627

  • Several genes using the three letter designation Sec or Yaj are presently annotated to GO:0015627. These need to be moved to “Sec complex” GO:0031522


type III protein secretion system complex ; GO:0030257

  • Change from being child of GO:0044459 “plasma membrane part” to child of GO:0044464 “cell part”

Synonyms

  • exact: TTSS complex
  • exact: T3SS complex
Jane

Done.


Sec complex (sensu Bacteria) ; GO:0031522

  • Change from being child of GO:0044459 “plasma membrane part” to child of GO:0044464 “cell part”

Definition:

  • A transmembrane protein complex involved in the translocation of proteins across the cytoplasmic membrane. In Gram-negative bacteria, Sec-translocated proteins are subsequently secreted via the type II, IV, or V secretion systems. Sec complex components include SecA,D,E,F,G,Y and YajC. As in, but not restricted to, the taxon Bacteria (Bacteria, ncbi_taxonomy_id:2).
Jane

Done.

Request to move mis-annotated genes out of GO:0015628

  • Several genes using the three letter designation Sec or Yaj are presently annotated to GO:0015628. These need to be moved to “Sec protein translocation system (sensu Bacteria)” (see above) when a GO term has been established


Proposed new terms

TAT protein translocation system complex (sensu Bacteria)

  • Child of “plasma membrane part” GO:0044459

Definition:

  • A complex of three proteins integral to the bacterial cytoplasmic membrane involved in translocation of folded proteins across the bacterial cytoplasmic membrane. As in, but not restricted to, the taxon Bacteria (Bacteria, ncbi_taxonomy_id:2).
Jane

Does this need to be a sensu term? Are there other TAT translocation systems outside of bacteria?.

Magdalen

Yes, the sensu term is required both here and in the coresponding biological process term, as there also exists a Tat translocation system in chloroplasts.


type IV protein secretion system complex

  • Child of “cell part” GO:0044464

Synonyms:

  • exact: T4SS complex

Definition:

  • A complex of proteins related to those involved in bacterial DNA conjugative transfer, that permits the transfer of nucleoprotein DNA conjugation intermediates or proteins into the extracellular milieu or directly into host cells. In general the type IV complex forms a multisubunit cell-envelope-spanning structure composed of a secretion channel and often a pilus or other surface filament or protein(s)
Jane

Done. GO id = GO:0043684

Pilus and fimbria terms in GO

Right now, GO has a number of fimbrial terms in the cellular component ontology.

There are two problems with this:

  1. It was decided that the term pilus should be primary to fimbria
  2. These terms are all children of intracellular organelle (GO:0043229) instead of cell projection part (GO:0044463) where they more logically belong (as a sibling of other projections like flagella)

I don’t know the proper protocol, but I favor:

  1. adding the two new terms below
  2. transferring annotation under fimbrium GO:0009289 to pilus part (below)
  3. obsoleting the following terms:
  • fimbrium GO:0009289
  • fimbrial shaft GO:0009418
  • fimbrial tip GO:0009419


Proposed new terms

pilus part

  • Child of “cell projection part” GO:0044463

Synonyms:

  • exact: fimbriae

Definition:

  • Any constituent part of a pilus, a short filamentous structure on the surface of bacterial cell variously involved in nucleic acid transfer, motility, and adherence to substrates


Type IV pilus part

  • Child of “pilus part” (see above)

Synonyms:

  • exact: type IV fimbriae
  • exact: TFP part
  • exact: type 4 pilus part

Definition:

  • Any constituent part of a Type IV pilus, a short filamentous structure on the surface of a bacterial cell distinguished from other pili by post-translational N-methylation of the pilin monomers
Jane

I suggest we do the following name changes:

  • fimbrium GO:0009289 -> pilus
  • fimbrial part GO:0044443 -> pilus part

and obsolete these two if you think we don't need them (what is the reason for obsoleting these by the way?):

  • fimbrial shaft GO:0009418
  • fimbrial tip GO:0009419

I can then add a new term 'type IV pilus' as a child of 'pilus'. We don't need a 'type IV pilus part' term, all proteins that are part of a type IV pilus can be directly annotated to 'type IV pilus' - the part terms are really just an artifact of making cell component is_a complete.

Magdalen

I agree that the easiest thing to do is not to create new pilus terms and obsolete fimbrial terms, but rather to keep the term numbers and just change the terminology (fimbrium to pilus). In that case, I have no problem with just changing fimbrial shaft and fimbrial tip to pilus shaft and pilus tip

Other questions

Can adhesion to host (GO:0044406) be a child of cell-substrate adhesion (GO 0031589)?

Amelia

I don't think so, unless an organism is a kind of substrate. Perhaps we should have a 'cell-organism surface adhesion' term?

Jane

I think this is now fixed - when we fixed up biological process, we made a new term 'biological adhesion' which adhesion to host (GO:0044406) and cell-substrate adhesion (GO 0031589) are both children of.

Marcus

Two other questions I came across while considering protein transport:

GO:0042953 lipoprotein transport is_a GO:0015031 protein transport. This is true because a lipoprotein is a type of protein. But… GO:0015869 DNA-protein complex transport is_a GO:0015031 protein transport, which begs the question, is a DNA-protein complex a protein?

Why is GO:0009306 protein secretion (“The regulated release of proteins from a cell or group of cells”) a sibling of GO:0015031 protein transport (“The directed movement of proteins into, out of, within or between cells”) Isn’t “regulated release” a type of “directed movement… into, out of or within or between”? Why is secretion not a child of transport?

Amelia

In answer to the second question, at some point in the near future we are going to be splitting up transport and secretion into cellular transport, cellular secretion, organismal transport and organismal secretion. Once those changes are made, secretion will be made a child of transport.



Trudy

The statements below are made in response to issues raised by Marcus and Magdalen on the PAMGO discussion forum.

Marcus 1. I am under the assumption that we even need this term [secretion terms under biological process] because we want to be able to annotate the protein being secreted, for example an “effector,” not just the protein components of the secretory complex. Otherwise why not just dual annotate to GO:0009306 protein secretion and one of the appropriate protein secretion system complex terms, e.g. GO:0030256?

Magdalen’s response: At present, one cannot annotate the proteins being secreted to the process term (something that some of us would really like to change!) However, even without annotating the secreted proteins to these process terms, I still favor creation of biological process terms for each secretion system. Afterall, they are distinct processes. On the other hand, maybe they don’t add much to the information represented by the corresponding cellular components terms. Anyone else have an opinion on this?


Trudy's response: I am beginning to think we are populating the ontology with so many terms that at some point users are going to be confused as to which terms to annotate their gene products to.

These are the various terms we have that one can likely annotate attributes of gene products associated with the secretory pathways to. ( egs. chaperones, translocators, effectors, structural proteins). I have listed them below for clarity.


(A) Terms associated with the secretory systems in the general ontology.

Type 1 protein secretor activity(0) F GO:0015428

Type 1 protein secretion system (2): being changed to “Protein secretion by the type I secretion system” [P] GO:0030253

Type 1 protein secretion complex (0) [C] GO:0030256


Type II protein secretor activity (0) [F] GO:0015447 Type 1I protein secretion system (59): being changed to “Protein secretion by the type II secretion system” [P] GO:0015628 Type II protein secretion system complex (38) [C] GO:0015627


Type III protein secretor activity (0) F GO:0015448 Type III protein secretion system (11) being changed to “Protein secretion by the type III secretion system” [P] GO:0030254 Type II protein secretion system complex (0) [C]GO:0030257


Type IV protein secretor activity (0) [F] GO:0015449 Type IV protein secretion system (0). Being changed to “Protein secretion by the type IV secretion system” [P] GO:0030255 No component term available


Type V protein secretor activity (2) [F] GO:0015323 Type V protein secretion system (2): GO:0030255 being changed to “Protein secretion by the type V secretion system “ No component term available.

Summary We have function, process and component terms for almost all the secretion terms from Type I to Type V.

All the function terms and also the Type II protein secretion system complex term have no definitions.

Among the function terms, the “Type I protein secretor activity’ is the only one that has one gene product annotated to it, all others have none Type II process term has the most annotations (59)


(B)Now under the IBO, the newly created terms relevant to this discussion are in bold

symbiosis, encompassing mutualism through parasitism

          interaction with other organism via secreted substance during symbiotic 
          interaction
                           interaction with host via secreted substance 
                                     interaction with host via protein secreted by type II secretion 
                                     system
                                     interaction with other organism via protein secreted by type 
                                     III secretion
                                     system during symbiotic interaction
                                     interaction with other organism via substance secreted by 
                                     type IV secretion
                                     system during symbiotic


The big question is do we need all these terms in the ontology. To answer this we would have to ask ourselves another question “what comprise the various secretion systems and under which terms can they be annotated. This is what I think from what I gather from the literature.

Quite easily, all the effectors (the gene products that are actually secreted to interact with the plant) can be captured under the new terms under the IBO. In eukaryotic microbes, at least with Phytophthora, I do not think the facilitators of the secretory process are considered secreted products interacting with the host. That may not be the case for bacteria.

All the structural gene products can be captured under the component terms .

What are the function terms capturing? Currently if you look at the numbers in parenthesis under (A) the most annotated to any function term is two. Can all the other gene products (besides the effectors directly interacting with the plant and the components of the systems) associated with the secretory systems be annotated to the function terms? If this is doable then we probably do not need the process terms in the general ontology.

Any further thoughts?????


Magdalen

Trudy, thanks for the excellent comments and questions.

I think your concern about too many terms is justified. A symptom of this may well be that I was completely unaware of molecular function terms pertaining to the secretion systems! Perhaps in part because “secretor activity” is not a phrase that is used by those doing research in bacterial secretion, and so did not find it in any of my searches of pre-existing GO terms. Furthermore, I have generally conceived of “molecular function” as a more specific, biochemically-oriented term. For example, within a secretion complex, protein “A” might proteolytically cleave protein “B”, and be appropriate for a molecular function term. In contrast, the “secretor” function carried out by the complex strikes me as more of a “biological process” than a “molecular function”. Consequently, I strongly advocate that for now, the molecular function terms for secretion systems I, II, III, IV and V be eliminated.

Regarding the “interaction with host…” terms: As things now stand, the only means of associating a secreted protein with the secretion complex is by use of the “interaction with host via protein secreted by type # secretion system”, and therefore I think these terms should be retained. However, given that these terms affect only to the secreted proteins and not the secretion systems themselves, this issue is somewhat distinct from the multiple overlapping terms applied to the secretion systems.

Returning to the secretion systems themselves and what terms for cellular component and biological process are most appropriate, I see two options:

(a) Cellular component = “type # secretion system complex” and Biological process = “protein secretion” (b) Cellular component = “type # secretion system complex” and Biological process = “protein secretion by the type # secretion complex”

As a protein secretion enthusiast, I tend toward option (b) because the precise nature of the biological process for each of the 5 systems is distinct, and can be spelled out in the definition. Furthermore, though there is precedence for describing protein complexes as cellular components, I see secretion as most fundamentally being a biological process (i.e. the most important ontology of the three when looking at “protein secretion”)

On the other hand, if one employed option (a), much of the differences in process could be specified in the “cellular component” definition leading to an overall result that is more condensed (i.e. fewer terms) and perhaps more in line with the original mission of GO.

In summary, I am advocating the following for protein secretion terms: Omit “secretor activity” Molecular Function terms Retain Cellular Component and Biological Process terms Retain “interaction with host via protein secreted by type # secretion system” as a means of capturing information about the secretion systems used by a given virulence protein. (Note that if we have difficulty coming to consensus I would rather go with option (a)and reserve the right to add more granular terms later, than to hold up the implementation of the overall body of protein secretion terms.)

Any other thoughts?