Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

is_a complete biological process ontology

Waclaw Kusnierczyk waku at idi.ntnu.no
Tue Jan 9 02:01:10 PST 2007


it seems, in the case of 'multicellular organismal process', 'cellular
process', and 'multiorganismal process', that there is a confusion:
- criteria for distinguishing categories are taken for
- categories themselves.

one could think of three (almost) orthogonal criteria here:

1. the count of cells in an (adult?) organism;
2. the extension of a process in terms of how many cells need to be
involved in an involved organism;
3. the extension of a process in terms of how many organisms need to be
involved.

(the criteria are _almost_ orthogonal in that there cannot be more than 
one cell involved within an organism if the organism is unicellular; 
thus 1. and 2. are partially dependent.)

the criterion 1. enables us to distinguish the following categories of
processes:
1a. those taking place in unicellular organisms,
1b. those taking place in multicellular organisms.
[1c. those taking place in acellular organisms.]

we should remember that some (most?) multicellular organisms begin their
lives as unicellular ones; this might complicate the issue if (the
stages of) development is considered.

the criterion 2. enables us to distinguish the following categories of
processes:
2a. those completely located within a single cell,
2b. those involving more than one cell.
[2c. those not taking place in a cell.]

though this criterion seems clear at face, it may not be easy to 
determine whether a process is 2a. or 2b.  this problem was already 
visible in the discussion below.

the criterion 3. enables us to distinguish the following categories of 
processes:
3a. those completely located within a single organism,
3b. those involving more than one organism.
[3c. those not involving any organisms.]

(is a process of viral infection 3a. or 3b.?)

the criteria are orthogonal, so that the categories they distinguish 
need not be disjoint.  whether the categories distinguished by a 
particular criterion are disjoint is another issue.  a particular 
process instance is either 1a. or 1b., but can it be both?  (development 
from a single cell into a multicellular organism?)  a particular type of 
processes may, though, have some of its instances 1a., while others 1b.
(in principle; examples?)

furthermore, a process that involves more than one organism may involve 
unicellular organisms and multicellular organisms at the same time.  it 
may thus be both 1a. and 1b., and may be both 2a. and 2b.

by combining these criteria, we may distinguish processes:

3a 1a 2a
3a 1b 2a
3a 1b 2b

3b 1a 2a
3b 1b 2a
3b 1b 2b
3b 1b 2a+2b
3b 1a+1b 2a
3b 1a+1b 2a+2b

since it gets complicated, it would perhaps be best to represent only 
categories distinguished by each criterion alone, and create multiple 
isa links to these categories, rather than explicitly represent all 
combinations.


to summarize:

- your 'multicellular organismal process' is 1b., but may be either 2a. 
or 2b. (or both, if 3b.), and may be either 3a. or 3b.

- your 'cellular process' is 2a., but may be either 1a. or 1b. (or both, 
if 3b.), and may be either 3a. or 3b.

- your 'multiorganism process' is 3b., and may be either 1a., 1b., or 
1a+1b., as well as 2a., 2b., or 2a.+2b.


does this make sense to you?

vQ




Wacek Kusnierczyk wrote:
> one more comment to 'multi-organismal process'.
> 
> if we consider viruses as organisms, then a multi-organismal process 
> must necessarily be either a multi-cellular process (if more than one 
> cell is involved) or a single-cellular process (in the case a cell 
> interacts with one or more viruses);  let's agree for the moment that 
> non-cellular organismal processes do not exist.
> 
> then a multi-organismal process does not have to be a multi-cellular 
> process, but it cannot be a non-multi-cellular, non-single-cellular 
> process anyway.  the three categories you claim to be disjoint cannot be 
> disjoint on any reasonable interpretation of the terms.
> 
> vQ
> 
> Wacek Kusnierczyk wrote:
>> an interesting initiative.
>> i made some inline comments.
>>
>>
>> Jennifer Clark wrote:
>>> Dear GO-Friends,
>>>
>>> Areas worked on:
>>> 1. Processes are now organized under the major categories of 
>>> multicellular organismal processes, cellular processes (including 
>>> single-celled organisms), and multi-organism processes (formerly 
>>> 'interaction between organisms'; there has been no change in meaning 
>>> of this term, only the name has changed). These categories are 
>>> disjoint; that is, no terms can have more than one of these terms as 
>>> an is_a parent.
>>
>> it is not clear what a multicellular organismal process is.  is it a
>> process that may take place only in multicellular organisms? (i.e., one
>> that cannot take place in a unicellular organism)
>>
>> if yes, then why the definition of the term 'multicellular organismal
>> process' is 'The biological processes, occurring at the level of the
>> organism, pertinent to the function of the organism', which covers many
>> processes taking place in unicellular organisms?
>>
>> if no, and if the term means a process that _may_ take place not only in
>> unicellular organisms, what is then the difference between this term and
>> the term 'cellular process' which 'includes single-celled organisms'
>> (rather than 'only single-celled organisms')?
>>
>> the latter is defined as 'Processes that are carried out at the cellular
>> level, but are not necessarily restricted to a single cell. For example,
>> cell communication occurs among more than one cell, but occurs at the
>> cellular level.'  why cannot such a process be a multi-organismal
>> process? (there can be communication between unicellular organisms;
>> would this not be covered by your definition?)
>>
>> there does not seem to be a clear ground for the claim that the three
>> (categories represented by the) terms you mention are disjoint.  perhaps
>>  the intention is that they are, but the definitions are hardly an
>> evidence for such an intention.
>>
>> in fact, if there were a clear cut ditinction between the terms
>> 'multicellular process' (only those processes that necessarily involve
>> more than one cell) and 'cellular process' (only those processes that
>> involve single cells, though may concurrently take place in many cells
>> as distinct processes), then a multi-organism process would necessarily
>> be a multicellular process.  i am not sure why communication that occurs
>>  among more than one cell is a (uni)cellular process, while cytokine
>> production is a multicellular process, although a (particular molecule
>> of a) cytokine is produced completely by one single cell.  if you decide
>> that the cell's being stimulated to the production is a good basis for
>> classifying the process as a multicellular process, why is not
>> communication (which is nothing else than sending and receiving stimuli)
>>  a multicellular process?
>>
>> (btw. i think you should avoid defining terms such as 'X process' as
>> 'the processes...'.  be consistent.)
>>



--
This message is from the GOFriends moderated mailing list.  A list of public
announcements and discussion of the Gene Ontology (GO) project.
Problems with the list?           E-mail: owner-gofriends at geneontology.org
Subscribing   send   "subscribe"   to   gofriends-request at geneontology.org
Unsubscribing send   "unsubscribe"  to  gofriends-request at geneontology.org
Web:          http://www.geneontology.org/



More information about the go-friends mailing list