antibody, vitamin and monomer

[Melanie Courtot]

Hi all,

The role branch that would like to remove some of the terms that we
think are not roles.

1. antibody

Following the long antibody discussion at http://sourceforge.net/tracker/?func=detail&aid=2415872&group_id=36855&atid=440764
, I would propose to mireot the term "GO:0042571, immunoglobulin
complex,
circulating." under protein complex. This term has antibody as exact
synonym.
See attached pic: we already imported protein complex from GO, so we
are not saying anything wrong by importing antibody as a subclass of
protein complex.

2. vitamin

There is a term vitamin in ChEBI, CHEBI:33229, see http://www.ebi.ac.uk/chebi/searchFreeText.do?searchString=CHEBI%3A33229
.
We could mireot it as well, under MaterialEntity.

3. monomer

This term has been bouncing between biomaterial, role and possibly
denrie. I would rather find it a stable home before moving it yet any
other place :)
In my opinion, monomer is some kind of organizational unit, so would
go under DENRIE, somehow related to data structure, except it would be
biological structure (with polymer, crystals...maybe?).

Wikipedia gives: http://en.wikipedia.org/wiki/Monomer : "A mononer is
a small molecule is a small molecule that may become chemically bonded
to other monomers to form a polymer. The most common natural monomer
is glucose, which is linked by glycosidic bonds into polymers such as
cellulose and starch, and is over 33% of the weight of all plant
matter."

If no objection to 1 and 2 I will update the file tomorrow. Your input
on 3 would be appreciated.

Thanks,
Melanie

[Alan Ruttenberg]

On Mon, Mar 30, 2009 at 6:45 PM, Melanie Courtot <mcou...@gmail.com> wrote:
> Hi all,
>
> The role branch that would like to remove some of the terms that we
> think are not roles.
>
> 1. antibody
>
> Following the long antibody discussion at http://sourceforge.net/tracker/?func=detail&aid=2415872&group_id=36855&atid=440764
> , I would propose to mireot the term "GO:0042571, immunoglobulin
> complex,
> circulating." under protein complex. This term has antibody as exact
> synonym.
> See attached pic: we already imported protein complex from GO, so we
> are not saying anything wrong by importing antibody as a subclass of
> protein complex.

Doesn't work for engineered "antibodies". However there is no problem
importing it and using it when appropriate (for denoting naturally
produced antibodies that are still part of their originating
organism).

> 2. vitamin
>
> There is a term vitamin in ChEBI, CHEBI:33229, see http://www.ebi.ac.uk/chebi/searchFreeText.do?searchString=CHEBI%3A33229
> .
> We could mireot it as well, under MaterialEntity.

It's defined as role. So MIREOT it under role. However report to CHEBI
that their current axiomitization is inconsistent, as roles can't be
part of materials.

Outgoing vitamin (CHEBI:33229) is a physiological role (CHEBI:52211)
Incoming food (CHEBI:33290) has part vitamin (CHEBI:33229)
(+)-α-tocopherol (CHEBI:18145) has role vitamin (CHEBI:33229)
fat-soluble vitamin (CHEBI:24020) is a vitamin (CHEBI:33229)
water-soluble vitamin (CHEBI:27314) is a vitamin (CHEBI:33229)

> 3. monomer
>
> This term has been bouncing between biomaterial, role and possibly
> denrie. I would rather find it a stable home before moving it yet any
> other place :)
> In my opinion, monomer is some kind of organizational unit, so would
> go under DENRIE, somehow related to data structure, except it would be
> biological structure (with polymer, crystals...maybe?).
>
> Wikipedia gives: http://en.wikipedia.org/wiki/Monomer : "A mononer is
> a small molecule is a small molecule that may become chemically bonded
> to other monomers to form a polymer. The most common natural monomer
> is glucose, which is linked by glycosidic bonds into polymers such as
> cellulose and starch, and is over 33% of the weight of all plant
> matter."
>
> If no objection to 1 and 2 I will update the file tomorrow. Your input
> on 3 would be appreciated.

I think I would like to see examples of intended usage on 3).

-Alan

>
> Thanks,
> Melanie
>
>
>
>
>
> >
>
>
> ---
> Mélanie Courtot
> TFL- BCCRC
> 675 West 10th Avenue
> Vancouver, BC
> V5Z 1L3, Canada
>
>
>
>
>
>

[Frank Gibson]

2009/3/31 Alan Ruttenberg <alanrut...@gmail.com>:

>
> On Mon, Mar 30, 2009 at 6:45 PM, Melanie Courtot <mcou...@gmail.com> wrote:
>> Hi all,
>>
>> The role branch that would like to remove some of the terms that we
>> think are not roles.
>>
>> 1. antibody
>>
>> Following the long antibody discussion at http://sourceforge.net/tracker/?func=detail&aid=2415872&group_id=36855&atid=440764
>> , I would propose to mireot the term "GO:0042571, immunoglobulin
>> complex,
>> circulating." under protein complex. This term has antibody as exact
>> synonym.
>> See attached pic: we already imported protein complex from GO, so we
>> are not saying anything wrong by importing antibody as a subclass of
>> protein complex.
>
> Doesn't work for engineered "antibodies". However there is no problem
> importing it and using it when appropriate (for denoting naturally
> produced antibodies that are still part of their originating
> organism).

I know this is copying the GO hierarchy but I think it is the
assertion under celluar component that casues the issues. I would
prefer if antibody is_a protein is_a material. I am not sure why
"complex" is important. Then we could have a relation to describe if a
particular instance of antibody was a cellular component or not. I
feel _cellular_component_ is a bit of a catch all term and maybe does
not have a place (out of scope) within OBI.

Frank

--
Frank Gibson, PhD
http://peanutbutter.wordpress.com/

[Fostel, Jennifer (NIH/NIEHS) [C]]

i can make an argument that monomer might be a role and a disposition.
if it is required, we can create a monomer role if we have a process of polymerization ro crystallization (but these are a natural chemical/physical processes outside the scope of OBI)and a biomaterial with disposition to polymerize (or possibly crystallize)

in reality, i see the use of monomer in OBI to describe a fait accompli -- the description of a polymer in hand by referencing it;s (formerly) monomeric units

...jennifer

[jie zheng]

1. antibody
I totally agree with Frank.

2. vitamin
I prefer to put it under MaterialEntity rather than Role. So, we can
MIREOT it from ChEBI. If it is not the exactly definition in OBI as in
ChEBI, we cannot MIREOT it. Just as we discussed before that FMA
"heart" is not the EOO "heart". We need two terms.

3. monomer
The term is better to go under DENRIE. It describes organization about
the molecular entity, monomer or polymer.

Jie

[Alan Ruttenberg]

On Tue, Mar 31, 2009 at 10:54 AM, jie zheng <zheng...@gmail.com> wrote:
>
> 1. antibody
> I totally agree with Frank.

Please see
http://sourceforge.net/tracker/index.php?func=detail&aid=2415872&group_id=36855&atid=440764

and see whether you feel the same way.

> 2. vitamin
> I prefer to put it under MaterialEntity rather than Role. So, we can
> MIREOT it from ChEBI. If it is not the exactly definition in OBI as in
> ChEBI, we cannot MIREOT it. Just as we discussed before that FMA
> "heart" is not the EOO "heart". We need two terms.

As I pointed out, ChEBI defines vitamin as a role.
http://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI%3A33229

One can always create a defined class for material e.g.

vitamin = material entity and bears vitamin role.

> 3. monomer
> The term is better to go under DENRIE. It describes organization about
> the molecular entity, monomer or polymer.

My gut is that there is more than information here, but I'm not able
to think how to say it now.

>
> Jie
> >
>

[jie zheng]

On Mar 31, 2:41 pm, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:

> On Tue, Mar 31, 2009 at 10:54 AM, jie zheng <zhengj2...@gmail.com> wrote:
>
> > 1. antibody
> > I totally agree with Frank.
>

> Please seehttp://sourceforge.net/tracker/index.php?func=detail&aid=2415872&grou...

>
> and see whether you feel the same way.

I will say that antibody is_a_celluar component bothers me. I agree
with Alan that the term imported from GO is naturally produced
antibody. I disagree that engineered "antibody" is_a_celluar
component.

> > 2. vitamin
> > I prefer to put it under MaterialEntity rather than Role. So, we can
> > MIREOT it from ChEBI. If it is not the exactly definition in OBI as in
> > ChEBI, we cannot MIREOT it. Just as we discussed before that FMA
> > "heart" is not the EOO "heart". We need two terms.

> As I pointed out, ChEBI defines vitamin as a role.http://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI%3A33229

>
> One can always create a defined class for material e.g.
>
> vitamin = material entity and bears vitamin role.

Sorry. When I first read Alan's message, I thought vatamin is defined
as a material in ChEBI. Since vitamin in ChEBI is a role, we'd MIREOT
it under Role rather than MaterialEntity.

Jie

[Frank Gibson]

On Tue, Mar 31, 2009 at 8:19 PM, jie zheng <zheng...@gmail.com> wrote:

On Mar 31, 2:41 pm, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:

> On Tue, Mar 31, 2009 at 10:54 AM, jie zheng <zhengj2...@gmail.com> wrote:
>
> > 1. antibody
> > I totally agree with Frank.
>

> Please seehttp://sourceforge.net/tracker/index.php?func=detail&aid=2415872&grou...

>
> and see whether you feel the same way.

I will say that antibody is_a_celluar component bothers me. I agree
with Alan that the term imported from GO is naturally produced
antibody. I disagree that engineered "antibody" is_a_celluar
component.

The complex issue is for a different thread :) within OBI we should talk about an antibody/protien being a material and determine if specific instances are specific cellular components or not. That way we are not constraining ourselves unnecessarily

Frank

 

> > 2. vitamin
> > I prefer to put it under MaterialEntity rather than Role. So, we can
> > MIREOT it from ChEBI. If it is not the exactly definition in OBI as in
> > ChEBI, we cannot MIREOT it. Just as we discussed before that FMA
> > "heart" is not the EOO "heart". We need two terms.
> As I pointed out, ChEBI defines vitamin as a role.http://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI%3A33229
>
> One can always create a defined class for material e.g.
>
> vitamin = material entity and bears vitamin role.

Sorry. When I first read Alan's message, I thought vatamin is defined
as a material in ChEBI. Since vitamin in ChEBI is a role, we'd MIREOT
it under Role rather than MaterialEntity.

Jie

[Alan Ruttenberg]

On Tue, Mar 31, 2009 at 3:19 PM, jie zheng <zheng...@gmail.com> wrote:
>
>
>
> On Mar 31, 2:41 pm, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:
>> On Tue, Mar 31, 2009 at 10:54 AM, jie zheng <zhengj2...@gmail.com> wrote:
>>
>> > 1. antibody
>> > I totally agree with Frank.
>>
>> Please seehttp://sourceforge.net/tracker/index.php?func=detail&aid=2415872&grou...
>>
>> and see whether you feel the same way.
>
> I will say that antibody is_a_celluar component bothers me. I agree
> with Alan that the term imported from GO is naturally produced
> antibody. I disagree that engineered "antibody" is_a_celluar
> component.

Right. So the question is, how do we record the commonality of
antibody in the sense of go and antibody in the sense of engineering.

Both have a common function, I would say. In addition, the ones we use
in experiments get a role.

-Alan

[Melanie Courtot]

On 31-Mar-09, at 1:07 PM, Alan Ruttenberg wrote:

>
> On Tue, Mar 31, 2009 at 3:19 PM, jie zheng <zheng...@gmail.com>
> wrote:
>>
>>
>>
>> On Mar 31, 2:41 pm, Alan Ruttenberg <alanruttenb...@gmail.com> wrote:
>>> On Tue, Mar 31, 2009 at 10:54 AM, jie zheng <zhengj2...@gmail.com>
>>> wrote:
>>>
>>>> 1. antibody
>>>> I totally agree with Frank.
>>>
>>> Please seehttp://sourceforge.net/tracker/index.php?func=detail&aid=2415872&grou
>>> ...
>>>
>>> and see whether you feel the same way.
>>
>> I will say that antibody is_a_celluar component bothers me. I agree
>> with Alan that the term imported from GO is naturally produced
>> antibody. I disagree that engineered "antibody" is_a_celluar
>> component.
>
> Right. So the question is, how do we record the commonality of
> antibody in the sense of go and antibody in the sense of engineering.
>
> Both have a common function, I would say.

> In addition, the ones we use
> in experiments get a role.

I'm not sure I understand: if we say antigen binding is the common
function, what would be the additional role borne by engineered
antibodies? Would it be for example a therapeutic role? (e.g. http://www.nature.com/nm/journal/v9/n1/full/nm0103-129.html)

I would expect to handle engineered antibody in the same way we will
deal with "extracted heart", i.e. a processed material (output of
processing process) having some kind of relation (derives_from or
else) to the "original" antibody. (Note that I just use an example
here, we haven't solved completely yet how to deal with EOO)

In order to move forward I would propose to mireot antibody for GO,
adding an editor note specifying "This class doesn't cover engineered
antibodies" and deal later on with engineered antibody.

>
>
> -Alan
>
>>
>>>> 2. vitamin
>>>> I prefer to put it under MaterialEntity rather than Role. So, we
>>>> can
>>>> MIREOT it from ChEBI. If it is not the exactly definition in OBI
>>>> as in
>>>> ChEBI, we cannot MIREOT it. Just as we discussed before that FMA
>>>> "heart" is not the EOO "heart". We need two terms.
>>> As I pointed out, ChEBI defines vitamin as a role.http://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI%3A33229
>>>
>>> One can always create a defined class for material e.g.
>>>
>>> vitamin = material entity and bears vitamin role.
>>
>> Sorry. When I first read Alan's message, I thought vatamin is defined
>> as a material in ChEBI. Since vitamin in ChEBI is a role, we'd MIREOT
>> it under Role rather than MaterialEntity.
>>

Ok - Jennifer is that ok with you?



Remaining "monomer" :)

(I didn't submit the term originally, so maybe somebody else will add
or correct)

I did find examples of usage like:
- http://pslc.ws/macrog/labnotes/monfilt.htm: "liquid monomers, such
as styrene, can be purified by distillation [..] here is a procedure
to purify monomer using a simple column"
- http://www.rpi.edu/dept/bcbp/molbiochem/MBWeb/mb2/part1/actin.htm :
actin monomers and polymerization
- http://www.plosone.org/article/info:doi/10.1371/journal.pone.
0003497: "Importantly, homodimeric mutant SOD1 forms unnatural,
partially folded monomeric and soluble oligomeric intermediates before
aggregation in vitro [11] and in vivo [12]."

This is why I mentioned that it seemed to me we are referring to a
specific configuration/organization when discussing monomer, polymers,
homodimers, heterodimers. Crystals are also important for protein
structure resolution (e.g. http://www.e-mep.org/), so they would be
covered by OBI too.

Thanks,
Melanie


>> Jie