Epistemic formalism (was Re: [Phenoscape] Re: [go] evidence code ontology)

[Ben Hitz]

> I think Tanya really hits the nail on the head here. I don't think  
> any of us exhaustively annotate NOT.

That is what I was trying to get at in my earlier email.

>  If my memory serves me correctly the original discussion of NOT  
> was for something like a peptidase that had great sequence  
> similarity to other known peptidases, but was shown to lack a key  
> residue and was therefore catalytically inactive.

But even this case is wrong!  There are kinases (and probably  
proteases as well) which are missing key catalytic residues that  
_are_ active (they use a slightly different catalytic mechanism, or  
have a slightly different fold)  And assertion of "inactivity" for an  
enzyme is flawed becase you never test all possible conditions and  
substrates.

The converse isn't true of positive annotations - someone says in  
their paper "I did Z and found that X did Y".   If X didn't do Y then  
it's true that "I did Z and found that X didn't do Y" is a valid  
piece of information... if the GO tracked Zs - but it doesn't.

Scientists use the GO as a set of assertions:  The GO asserts that  
(under some undefined conditions) that gene_product X does Y.  The  
assertion that X doesn't do Y is much less useful when you don't know  
what conditions it's under.

Ben
--
Ben Hitz
Senior Scientific Programmer ** Saccharomyces Genome Database ** GO  
Consortium
Stanford University ** hitz at genome.stanford.edu