gocentral
commented
16 years ago Logged In: YES user_id=579762 Originator: YES
What we did after discussions following comments from Eurie and Val and after a Skype/Webex session with Kimberly.
1) Merged the 'regulation of progression through cell cycle' into 'regulation of cell cycle.'
2) Renamed the 'regulation of progression through' terms to their simpler forms when the simpler parents do not exist since the parent non-regulates terms are already defined as the progression of... As an example, 'regulation of progression through S phase' was renamed as 'regulation of S phase.'
(Note: Points 3 and 4 of our original proposal were NOT implemented.)
-
Created a new term 'cell cycle switching' to be a parent for the 'cell cycle switching, meiotic to mitotic phase' and 'cell cycle switching, mitotic to meiotic phase.' Placed this new term as a direct child of 'cell cycle process.'
-
Removed 'cell cycle process' as an is_a parent of 'regulation of cell cycle.' The latter term is now only an is_a child of 'regulation of cellular process' and a part_of (soon to be regulates) child of 'cell cycle.'
Original comment by: tberardini
For archive purposes, I'm reproducing the email based discussion.
Original email:
Dear GO Consortium,
As part of the 'regulates' project, we have come across a set of terms in the cell cycle portion of the graph that seem problematic. We are writing because we know that a lot of previous work has gone into this part of the graph and want to be sure that our plan makes sense. We have two sets of terms similar to these:
term: regulation of cell cycle (simple term) def:A cell cycle process that modulates the rate, extent or mode of the cell cycle.
term: regulation of progression through cell cycle (progression term) def: Any process that modulates the rate or extent of progression through the cell cycle.
Our question is how does 'regulation of progression through cell cycle' differ from 'regulation of cell cycle'?
Note that 'cell cycle' itself is defined as "the progression of ...phases and events ...", implying that regulation of the cell cycle is regulation of progression through the cycle; the same holds for the types of cell cycle (i.e. its is_a children).
cell cycle: The progression of biochemical and morphological phases and events that occur in a cell during successive cell replication or nuclear replication events. Canonically, the cell cycle comprises the replication and segregation of genetic material followed by the division of the cell, but in endocycles or syncytial cells nuclear replication or nuclear division may not be followed by cell division.
If they are indeed describing the same processes, the 'regulation of %' and 'regulation of progression of %' terms should be merged. If not, then we need to create a bona-fide biological process called 'progression through cell cycle' and we need to differentiate that from 'cell cycle' with a really good definition. Currently the progression terms are only used in conjunction with regulation.
The way we represent these in the graph is also inconsistent. An example follows:
Case 1: No direct link between the 'regulation of progression through cell cycle' term and 'cell cycle' term exists.
regulation of cell cycle --[i]regulation of progression through cell cycle
Case 2: Direct link between 'regulation of progression through meiotic cell cycle' and 'meiotic cell cycle' exists.
regulation of meiotic cell cycle --[i]regulation of progression through meiotic cell cycle
meiotic cell cycle --[p]regulation of progression through meiotic cell cycle
We propose:
1) Merging the 'regulation of progression through' terms with their simpler parents when they exist.
2) Renaming the 'regulation of progression through' terms to their simpler forms when the simpler parents do not exist since the parent non-regulates terms are already defined as the progression of...
3) Moving 'cell cycle switching, meiotic to mitotic cell cycle' to become an is_a 'negative regulation of meiotic cell cycle' and an is_a 'positive regulation of mitotic cell cycle'.
4) Moving 'cell cycle switching, mitotic to meiotic cell cycle' to be an is_a 'positive regulation of meiotic cell cycle' and an is_a 'negative regulation of mitotic cell cycle'.
We think that the last two terms were why 'regulation of cell cycle' and 'regulation of progression through cell cycle' were originally created as distinct terms. However, the movement of these as in 3 and 4 makes more logical sense since the process they describe stops (negatively regulates) one type of cell cycle and initiates (positively regulates) another.
Please comment if you think that these merges/renames are not ok. If this is the case then we will need to make the missing 'progression' terms or 'simpler terms' and will need help in differentiating the 'progression' terms from their parents.
David & Tanya
Eurie's comments:
I have a couple of questions -
Are the "regulation of cell cycle" and "regulation of progression through cell cycle" going to be specific for just regulating the rate at which the progression through the cell cycle occurs? Then maybe they should be renamed "regulation of cell cycle phase" (cell cycle phase is GO:0022403)?
Or do we need to keep the structure of the regulation terms to mimic how the cell cycle branch is organized?
"regulation of cell cycle" - general grouping term for both molecular/cellular events and temporal phasing -- "regulation of cell cycle processes" - grouping term for all the regulation of processes ---- "regulation of cell cycle phase" - specific for going through the temporal phases
The mitotic - meiotic switch terms sound ok. As long as there are synonyms to find them - most folks in the field do refer to "entry into meiosis" but there are positive and negative regulators of the process.
Val's comments:
At first glance, I think that what you propsed is OK. You are correct that the seperate terms were created to cope with the switching between mitotic and meiotic cell-cycles, but your solution of making these positive and negative regulatory terms respectively seems logically consistant.
I think however, there is a more fundamental problem in this portion of the graph in that there is some confusion between the concurrent use of meiosis to describe 'meiotic division' and meiosis as a 'developmental stage'
I'm not sure if the terms created to cope with the switching between mitotic and meiotic cell-cycles rather refer to a switch between developmental stage for example "entry into meiosis" for yeast would include response to pheromone, and conjugation etc, and gene products annotated to these terms would not necessarily be considered be a 'cell-cycle ' genes.
current child terms of "meiosis" also include megasporogenesis (synonym megaspore development ) and these terms do not appear to be referring specifically to meiosis as a cell cycle process.
This is essentially what I was trying to describe in SF 1831804 ] yeast meiotic development.
Even some of the definitions of these terms seem to merge both concepts into a single term.
It might make sense to tackle both issues at the same time. I'm not sure what the solution is....
David/Tanya's response:
Answers are interspersed below:
Eurie Hong wrote: > I have a couple of questions - > > Are the "regulation of cell cycle" and "regulation of progression through cell cycle" going to be specific for just regulating the rate at which the progression through the cell cycle occurs?
Not just the rate. The standard definition of 'regulation of xxx' is:
Any process that modulates the frequency, rate or extent of process, [definition of process].
Since the cell cycle is defined as the progression... then the regulation of the cell cycle would be anything that modulates the frequency, rate or extent of the progression.
Then maybe they > should be renamed "regulation of cell cycle phase" (cell cycle phase is GO:0022403)?
No, that would be a different term, because there are cell cycle phase terms in the ontology (see all children of 'cell cycle phase'). So the regulation of a cell cycle phase would describe the regulation of one of these, for example, 'regulation of interphase' (which doesn't yet exist).
> > Or do we need to keep the structure of the regulation terms to mimic how the cell cycle branch is organized? > > "regulation of cell cycle" - general grouping term for both molecular/cellular events and temporal phasing > -- "regulation of cell cycle processes" - grouping term for all the regulation of processes > ---- "regulation of cell cycle phase" - specific for going through the temporal phases
Yes. Our goal is to create regulates terms that don't conflict with the structure of the process ontology. Right now, we're not creating all of these terms but if and when they are needed, we should be able to insert them easily because the structure already exists.
> The mitotic - meiotic switch terms sound ok. As long as there are synonyms to find them - most folks in the field do refer to "entry into meiosis" but there are positive and negative regulators of the process.
Since we're not proposing changing the terms, synonyms or textual definitions but rather their locations in the graph, if there are synonyms that are missing, please let us know.
Now, on to Val's comments.
> On Dec 6, 2007, at 6:25 AM, Valerie Wood wrote: >
>> At first glance, I think that what you propsed is OK. You are correct that the seperate terms were created to cope with the switching between mitotic and meiotic cell-cycles, but your solution of making these positive and negative regulatory terms respectively seems logically consistant.
Great! We assume this means that our proposed changes 1-4 are ok.
>> >> I think however, there is a more fundamental problem in this portion of the graph in that there is some confusion between the concurrent use of meiosis to describe 'meiotic division' and meiosis as a 'developmental stage'
From its current placement in the graph, meiosis is a type of cell cycle phase. It is not restricted to simply meiotic division.
>> I'm not sure if the terms created to cope with the switching between mitotic and meiotic cell-cycles rather refer to a switch between developmental stage >> for example "entry into meiosis" for yeast would include response to pheromone, and conjugation etc, and gene products annotated to these terms would not necessarily be considered be a 'cell-cycle ' genes.
It seems that a problem here is that 'regulation of cell cycle' is_a 'cell cycle process.' If we remove this relationship, then 'regulation of cell cycle' will only have a regulates relationship with 'cell cycle' and will not be a cell cycle process.
If we adopt our proposal and remove this relationship, genes annotated to the terms 'cell cycle switching, mitotic to meiotic cell cycle' and 'cell cycle switching, meiotic to mitotic cell cycle' will be involved in regulation of the cell cycle and will not necessarily be annotated to terms that are part_of or is_a children of 'cell cycle.'
>> current child terms of "meiosis" also include >> megasporogenesis (synonym megaspore development ) and these terms do not appear to be referring specifically to meiosis as a cell cycle process.
Megasporogenesis, despite the name, does not refer to the genesis of the megaspore but specifically to the meiosis of the diploid megasporocyte to give rise to four haploid megaspores.
>> >> This is essentially what I was trying to describe in >> SF 1831804 ] yeast meiotic development. >> >> Even some of the definitions of these terms seem to merge both concepts into a single term. >> >> It might make sense to tackle both issues at the same time. I'm not sure what the solution is....
While a valid concern, this is not immediately relevant to our current project of cleaning up the ontology for implementation of the regulates relationship. We will address this issue at a later time.
We are still waiting to hear back from Kimberly who is also looking at their annotations to these terms. If no other conflicts arise, we look forward to making the changes on Dec. 10th.
Thank you both for your comments. Please let us know if we didn't address your issues satisfactorily.
Tanya and David
Kimberly's comments:
In the course of going through our annotations, I've come across cases that doesn't seem to work for change #4, if I understand it correctly.
4) Moving 'cell cycle switching, mitotic to meiotic cell cycle' to be an is_a 'positive regulation of meiotic cell cycle' and an is_a 'negative regulation of mitotic cell cycle'. We think that the last two terms were why 'regulation of cell cycle' and 'regulation of progression through cell cycle' were originally created as distinct terms. However, the movement of these as in 3 and 4 makes more logical sense since the process they describe stops (negatively regulates) one type of cell cycle and initiates (positively regulates) another.
As an example, we have a gene, glp-1, that regulates the decision between mitosis and meiosis in the germ line. glp-1 could be annotated to 'germline cell cycle switching, mitotic to meiotic cell cycle', but its role in this process is as a positive regulator of the mitotic cell cycle and a negative regulator of the meiotic cell cycle. This is opposite to what is proposed above.
One alternative is to make the cell cycle switching terms is_a children of regulation of mitotic cell cycle and regulation of meiotic cell cycle, but then we lose the directionality of the regulation. In that case we'd could additionally annotate to the respective positive and negative regulation terms that could also be is_a children of the regulation parent. Or, we could make positive and negative regulation terms for cell cycle switching to capture the appropriate directionality. glp-1 would then be annotated to negative regulation of germline cell cycle switching, mitotic to meiotic cell cycle. I would also then annotate glp-1 to positive regulation of cell proliferation, which is not related to the cell cycle terms in GO, I believe.
Val's 2nd comment:
I was thinking about this some more and I think we did need the 2 terms
regulation of cell cycle
and regulation of progression through cell cycle.
where regulation of cell cycle included regulation of the 'mode' of cell cycle and the regulators for switching between types of cycle.
Regulation of progression only includes the events after 'cell cycle' start (i.e after commitment to the type of cell cycle) and only includes the cyclical events and canonical cell cycle regulators and checkpoints. Otherwise there is no process term that can be used to retreive these cell cycle regulators.
David's response:
From the current structure of the graph it seems like we will need a term 'regulation of cell cycle phase', which Eurie hinted at in her original comments. The problem with 'regulation of progression' and just 'regulation' is that they mean exactly the same thing since the cycle itself is defined as a progression. Also remember, once the regulates relationship is in place, the regulation terms will not necessarily be is_a or part_of the parents themselves. We are going to have a look at the regulation of the modes of the cell cycle this afternoon. Kimberly had some really good points about these terms with respect to how it happens in worms and gene products do not necessarily act in the ways that they would first seem.
Reported by: tberardini
Original Ticket: "geneontology/ontology-requests/4640":https://sourceforge.net/p/geneontology/ontology-requests/4640