Re: Rules (was Re: Ambiguous names. was: Re: URL +1, LSID -1)

On Jul 17, 2007, at 1:44 AM, Eric Jain wrote:

> Chris Mungall wrote:
>> We have also switched from talk of defining specific proteins to  
>> rules to automatically annotate protein records.
>
> You're right, small digression, hope it's of interest anyway :-)

Definitely - although I don't think OWL/SWRL is quite the right tool  
for this job yet

although perhaps getting closer:

Putting OWL in Order:   Patterns for Sequences in OWL
Nick Drummond1, Alan Rector1, Robert Stevens1, Georgina M2Matthew  
Horridge1, Hai H. Wang1, Julian Seidenberg1
http://owl-workshop.man.ac.uk/acceptedLong/submission_12.pdf

(I'm not a big fan of the modeling biological sequences as lists  
approach, but I think the results could be replicated using a realist  
representation)

I still don't think the HMM gurus have much to worry about yet

>> I read "broad classes of proteins" as being more inclusive than  
>> the class denoted by OPSD_HUMAN in my interpretation, but also  
>> including for example all human opsin proteins, all vertebrate  
>> opsins, ...
>
> At that level of broadness I guess we're talking about protein  
> families and domains, which is above the abstraction level used in  
> UniProt. But for some people UniProt is already too broad, so I  
> guess it's a point of view thing.

yes it seems I was misinterpreting Alan here, so we are still talking  
at the sequence-per-species level

>> To summarise: the hypothesis is that any UniProt entry can be  
>> formally defined using OWL-DL in an automated fashion in a way  
>> that is reasonably concordant with the intent of UniProt. There  
>> may well be counter-examples that disprove this.
>
> One thing that may be worth pointing out here is that when  
> biologists have some sequences and want to see if there is any  
> relevant information in UniProt, they'll blast them, and look at  
> the best matches (subjective).

This would seem to be consistent with the hypothesis

> The one thing they'll certainly not do is use an inference engine :-)

There's an interesting challenge. Sequence alignment in Pellet anyone?

Received on Wednesday, 18 July 2007 06:16:32 UTC