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    

Draft evidence code ontology

Harold Drabkin hjd at informatics.jax.org
Mon Mar 1 07:32:12 PST 2004


I'm really having a hard time not thinking of this as a direct assay; it's
really a type of "coupled" assay; you couple the binding event to a
transcription event.  The fact that it's done in vivo rather than in vitro
shouldn't be the issue.
There are many types of enzyme assays that would be called "direct" but they
are actually rigged to measure one  product of a reaction by performing
another reaction on the product to measure it (e.g, 3-phospho-D-glycerate
kinase and glyceraldehyde-3-phosphate dehydrogenase ) If done in a crude
extract as opposed to a purified protein could bring in false
positive/negative aspects as well.


Another example:
I transfect a cell with two CAT reporters differing in a single amino acid,
and make a crude extract, and measure CAT level by an enzymatic assay. Cat
level is higher in one that the other. Does the mutation effect protein
activity?.. Maybe... But what if I did a western and a northern and found
that the difference in activity level was actually due to less protein
(Stability?) or mRNA (RNA stability?).

I would think an indirect assay might be one of the types where one treats
cells with a protein 1 f, and then sees the level of a particular protein 2
increase, and one annnotates the factor to biosynthesis of the protein 2
that you measured (when in fact it could be the factor decreased the
degradation of protein 2's mRNA, protein, secretion, etc.).

I suppose it could be a question of how many steps come in between what you
are exactly assaying, and what you say it's linked to.

> 
On 2/27/04 6:18 PM, "Karen Christie" <kchris at genome.stanford.edu> wrote:

> Hi,
> 
> comments inserted below
> 
> 
> On Fri, 27 Feb 2004, Harold Drabkin wrote:
> 
>> Forgot to add something:
>> I wouldn't call it a strong assay, but it is an assay none-the-less.
> 
> I'm not objecting to calling it an assay. I agree with you there that it
> is an assay, just not a direct one. Thus my actual argument was that the
> IPI code (which is the one that includes 2-hybrid experiments) should NOT
> map up into the IDA code in the way that Michael had proposed.

>> I don't think the ontology (if we go this route; remember it was voted down
>> a few go meetings ago) has to imply whether it's better than other type of
>> assay. It's just a type of assay.
> 
> I also completely agree with you there. I don't think GO evidence codes
> should attempt to denote quality, and they certainly don't do this right
> now.
> 
>> Perhaps we need 
>> 
>> Inferered from Experiment
>> Then Direct Assay
>> Then Indirect Assay
> 
> The direct/indirect split might be tricky. Certainly for IPI, some of the
> specific assays listed are direct, while others, e.g. 2-hybrid, are not.
> 
> -Karen
> 
>> ??
>> On 2/27/04 3:14 PM, "Karen Christie" <kchris at genome.stanford.edu> wrote:
>> 
>>> Well, personally, I don't think I'd always want IPI to map up into
>>> IDA, with respect to GO evidence codes.
>>> 
>>> For example, I don't think I'd consider 2-hybrid evidence to be IDA. While
>>> it is true that you know what plasmids you put into the cell, you do not
>>> know exactly what is binding. The presumption is generally that the target
>>> portion (let's call it T) of the activator construct is binding to the
>>> bait portion (let's call it B) of the DNA-binding construct, but this is
>>> indirect. You have not actually done a direct assay of the proteins T and
>>> B to determine that they interact directly with each other.
>>> 
>>> -Karen
>>> 
>>> 
>>> IDA inferred from direct assay
>>> Enzyme assays 
>>> In vitro reconstitution (e.g. transcription)
>>> Immunofluorescence (for cellular component)
>>> Cell fractionation (for cellular component)
>>> Physical interaction/binding assay (sometimes appropriate for cellular
>>> component or molecular function)
>>> 
>>> 
>>> IPI inferred from physical interaction
>>> 2-hybrid interactions
>>> Co-purification
>>> Co-immunoprecipitation
>>> Ion/protein binding experiments
>>> 
>>> 
>>> On Fri, 27 Feb 2004, Harold Drabkin wrote:
>>> 
>>>> Why wouldn't an experiment that fell under an IPI be a type of IDA?
>>>> 
>>>> On 2/27/04 11:23 AM, "Midori Harris" <midori at ebi.ac.uk> wrote:
>>>> 
>>>>> 
>>>>>>   %IDA
>>>>>>    %IPI etc
>>>>> 
>>>>> This still does not work for the specialized case of GO annotation.
>>>>> 
>>>>> m
>>>>> 
>>>>> 
>>>>> --
>>>>> 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/
>>>> 
>>>> 
>>>> --
>>>> 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/
>>>> 
>>> 
>>> 
>>> 
>>> --
>>> 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/
>> 
> 
> 
> 
> --
> 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/


--
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