reorganize terms under blood coagulation and add new terms

[gocentral]

GO:0007596 'blood coagulation' now has a definition 'The sequential process by which the multiple coagulation factors of the blood interact, ultimately resulting in the formation of an insoluble fibrin clot; it may be divided into three stages: stage 1, the formation of intrinsic and extrinsic prothrombin converting principle; stage 2, the formation of thrombin; stage 3, the formation of stable fibrin polymers.' and children: is_a GO:0007598: blood coagulation, extrinsic pathway is_a GO:0007597: blood coagulation, intrinsic pathway part_of GO:0030168: platelet activation

There are some problems with this, at least for human blood coagulation: a term is needed that corresponds to stages 2 and 3 of the definition, and all of these terms should be part_of children or grandchildren of blood coagulation (e.g., PMID:7598447), like this:

blood coagulation GO:0007596 .. p platelet activation GO:0030168 .. p fibrin clot formation GO:new .... p extrinsic pathway GO:0007598 .... p intrinsic pathway GO:0007597 .... p common pathway GO:new

Here are brief descriptions of these processes as they occur in humans; they show why the three pathways are parts rather than instances and possible definitions may be hidden in there someplace.

The extrinsic pathway is initiated when tissue factor (TF), an intrinsic plasma membrane protein, is exposed to the blood by injury to the wall of a blood vessel. TF is then able to bind factor VIIa from plasma, and possibly also factor VII, to form complexes capable of catalyzing the conversion of factor X, from plasma, into its activated form, factor Xa. Factor Xa catalyzes the conversion of additional factor VII molecules to their activated form, increasing the amount of tissue factor:factor VIIa complex available at the site of injury, accelerating the generation of factor Xa, and allowing the activation of factor IXa as well. This process is self-limiting because as levels of factor Xa increase, tissue factor:factor VIIa complexes become trapped in the form of catalytically inactive heterotetramers with factor Xa and the protein TFPI (tissue pathway factor inhibitor).

The intrinsic pathway of blood clotting connects interactions among high molecular weight kininogen, prekallikrein, and factor XII to the activation of clotting factor X by a series of reactions that is independent of the extrinsic pathway and that is not subject to inhibition by TFPI. It is essential for the prolongation of the clotting cascade: while the reactions of the extrinsic pathway appear to be sufficient to initiate clot formation, those of the intrinsic pathway are required to maintain it.

The common pathway consists of the cascade of activation events leading from the formation of activated factor X to the formation of active thrombin, the cleavage of fibrinogen by thrombin, and the formation of cleaved fibrin into a stable multimeric, cross-linked complex.

Peter D'E Reactome

Reported by: deustp01

Original Ticket: "geneontology/ontology-requests/4835":https://sourceforge.net/p/geneontology/ontology-requests/4835

[gocentral]

Logged In: YES user_id=865072 Originator: NO

Peter,

I think this is a sensible rearrangement of and addition to these terms. However, it would be great if you could suggest some (new) is_a parents for fibrin clot formation GO:new, blood coagulation, extrinsic pathway GO:0007598, blood coagulation, intrinsic pathway GO:0007597, and blood coagulation, common pathway GO:new. The existing terms GO:0007597 and GO:0007598 had blood coagulation GO:0007596 as their is_a parent.

Also, I think it is best to retain the full existing names for GO:0007597 and GO:0007598, including the "blood coagulation" part, to distinguish these terms clearly from any other pathway terms in the GO. And I'm sure the GO editors would love it if you could state explicit definitions for the terms.

Thanks,

Alex

Original comment by: addiehl

[gocentral]

Logged In: YES user_id=1101528 Originator: YES

Yes, keep 'blood coagulation' in the names:

blood coagulation GO:0007596 .. p blood coagulation, platelet activation GO:0030168 .. p blood coagulation, fibrin clot formation GO:new .... p blood coagulation, extrinsic pathway GO:0007598 .... p blood coagulation, intrinsic pathway GO:0007597 .... p blood coagulation, common pathway GO:new

and perhaps also add it to the 'platelet activation' term, in case there are processes in which platelet activation does not lead to blood clot formation.

The reason to make existing terms 'blood coagulation, extrinsic pathway' (GO:0007598) and 'blood coagulation, intrinsic pathway' (GO:0007597) (and also the new terms for fibrin clot formation and common pathway) part_of children of blood coagulation is that, in humans at least, these are all required parts of a single overall process, not optional alternative ways of doing it. At least, that appears to be the current expert consensus view (and also appears to be a change from the expert consensus view of 20 or 30 years ago, which was closer to the 'optional alternatives' view).

These four processes, 'extrinsic', 'intrinsic', 'common', and 'fibrin clot', are all examples of signal amplification by means of an enzyme cascade. Probably out of ignorance, I can't find any GO biological_process term like that to link them to.

Here are proposed definitions for the new terms, and also proposed revised definitions for some existing ones, to try to describe the process as it occurs in vivo, rather than as it occurs in vitro.

'blood coagulation, extrinsic pathway' The self-limited process linking exposure and activation of tissue factor to the activation of clotting factor X.

'blood coagulation, intrinsic pathway' The interactions among high molecular weight kininogen, prekallikrein, and factor XII that lead to the activation of clotting factor X.

'blood coagulation, common pathway' The cascade of enzyme activation events leading from the formation of activated factor X to the formation of active thrombin, the cleavage of fibrinogen by thrombin, and the formation of cleaved fibrin into a stable multimeric, cross-linked complex.

'blood coagulation, fibrin clot formation' The cascade of enzymatic reactions initiated by physical damage to the wall of a blood vessel, that leads to the formation of a formation of a fibrin clot at the site of the injury. The process also includes numerous positive and negative regulatory events.

Peter

Original comment by: deustp01

[gocentral]

Logged In: YES user_id=865072 Originator: NO

Peter and Midori,

In thinking about the parents for the blood coagulation pathway terms, I see that a general ‘protein activation cascade’ term could provide an is_a parent to these terms and a couple of other terms and serve as a useful grouping term for all these terms

New term: protein activation cascade Synonym: protein activation pathway Definition: A sequential series of modifications to a set of proteins where the product of one reaction acts catalytically in the following reaction. The modifications typically involve proteolysis or binding steps. is_a: GO:0032501 ! multicellular organismal process is_a: GO:0050896 ! response to stimulus

I am providing a definition, although it is not perfect, as I’m not sure binding is considered a catalytic event or even a modification in the GO sense. We could skip the word ‘catalytically’ or replace ‘acts catalytically’ with ‘potentiates’.

The term ‘protein activation cascade’ itself is an is_a to ‘multicellular organismal GO:0032501 process’ and ‘response to stimulus ; GO:0050896’. I think these are accurate relationships, as these cascades typically occur in extracellular space within a multicellular organism, but ‘cellular process ; GO:0009987’ might also represent a valid parent instead of ‘multicellular organismal process’, depending on how one interprets the definition of ‘cellular process’.

Terms which should have ‘protein activation cascade’ as an is_a parent:

complement activation ; GO:0006956 kinin cascade ; GO:0002254 blood coagulation, extrinsic pathway ; GO:0007598 blood coagulation, intrinsic pathway ; GO:0007597 blood coagulation, common pathway ; GO:new

The new is_a relationship for ‘complement activation ; GO:0006956’ indicated above would replace its current is_a relationship to ‘activation of plasma proteins during acute inflammatory response ; GO:0002541’ which was never quite right. The other is_a relationships for GO:0006956 should remain intact. Also, ‘complement activation ; GO:0006956’ should gain the exact synonym ‘complement cascade,’ and its children, ‘complement activation, alternative pathway’, ‘complement activation, classical pathway’, and ‘complement activation, lectin pathway’ should gain exact synonyms with ‘complement activation’ replaced by ‘complement cascade’.

I would also propose that all regulation terms be immediately created for ‘protein activation cascade’ as well as for ‘blood coagulation, extrinsic pathway’, ‘blood coagulation, intrinsic pathway’, and ‘blood coagulation, common pathway’, and linked to the higher level regulation terms (link the regulation terms that already exist for the complement and kinin terms as well).

One general consideration for the BP ontology is that while not all signal transduction processes rely on protein activation cascades, there are many that do, and consequently we may need to evaluate whether ‘protein activation cascade’ should have additional children. My impulse at the moment is to ignore this possibility for the moment while we think about it.

And, of course, I am open to discussion on all of this. Sorry about the amount of work this represents to you Midori, but that’s par for the course with me I’m afraid.

Thanks,

Alex

Original comment by: addiehl

[gocentral]

Logged In: YES user_id=1101528 Originator: YES

For the groups of extracellular cascades listed, the definitions and parent-child relationships all look good. In GO molecular_function and also in the minds of the real biochemists I asked, 'binding' function is separate from 'catalysis'. Here, though, that's not a problem because all of the cascades involve catalytic amplification: a small number of molecules of a proenzyme early in the process become active and catalyze the activation of a larger number of second-step enzyme molecules, which in turn catalyze the activation of still more ... and so on. Some of the activation events also involve binding (two inactive proteins associate to form an active complex) or unbinding (a repressive subunit dissociates, allowing an enzyme to become active), but the defining feature, I think, remains catalysis.

A thought for the future: there are numbers of intracellular cascades that also fit this model.

Peter

Original comment by: deustp01

[gocentral]

Logged In: YES user_id=865072 Originator: NO

Midori,

David Hill mentioned your mutual reluctance to use the word "activation" in my suggested term name, due to the use of "activation" in terms that are children of "positive regulation of X" terms. Although I find this a bit picky, since in fact the use in the term name I suggested matches one of the biologically accepted meanings of "activation", I suppose in order to enable Peter's useful changes to the coagulation pathways to be incorporated into the GO, we could use the alternative name, "protein activitory cascade," and keep my suggested term name as a synonym. Nevertheless, I like my original term name better, since "activitory" is a bit jargony, and hope that the two of you might come around to it.

Thanks,

Alex

Original comment by: addiehl

[gocentral]

I have, at long last, implemented changes as described in Alex's and Peter's comments of 2008-02-18 15:50:03, 2008-02-19 15:46:07 and 2008-02-20 16:49:22.

New terms: protein activation cascade GO:0072376 blood coagulation, common pathway GO:0072377 blood coagulation, fibrin clot formation GO:0072378

Regulation terms to come, when the above will be available for TermGenie (will close this item when regulation terms are live).

Note that I have made 'protein activation cascade' is_a 'response to stimulus' as Alex recommended, and also is_a protein metabolic process. I have not given it any other is_a parents because the comments here indicate that there are cellular- and multicellular-organismal-level cascades.

(I've satisfied myself that the concern about using "activation" univocally, though still valid, should not further impede this work, because we're already using a rather broad interpretation of activation -- it's actually broader than the usage in the regulation branch implies, and may eventually be cleaned up throughout the ontology, not just for activation cascades.)

m

Original comment by: mah11

[gocentral]

• assigned_to: nobody --> gomidori

Original comment by: mah11

[gocentral]

These are now in the TermGenie pipeline, and I'll close this item when they're live:

id: GO:2000257 name: regulation of protein activation cascade

id: GO:2000258 name: negative regulation of protein activation cascade

id: GO:2000259 name: positive regulation of protein activation cascade

id: GO:2000260 name: regulation of blood coagulation, common pathway

id: GO:2000261 name: negative regulation of blood coagulation, common pathway

id: GO:2000262 name: positive regulation of blood coagulation, common pathway

id: GO:2000263 name: regulation of blood coagulation, extrinsic pathway

id: GO:2000264 name: negative regulation of blood coagulation, extrinsic pathway

id: GO:2000265 name: positive regulation of blood coagulation, extrinsic pathway

id: GO:2000266 name: regulation of blood coagulation, intrinsic pathway

id: GO:2000267 name: negative regulation of blood coagulation, intrinsic pathway

id: GO:2000268 name: positive regulation of blood coagulation, intrinsic pathway

Original comment by: mah11

[gocentral]

• status: open --> closed-accepted

Original comment by: mah11