[go] AmiGO beta announcement

[Alexander Diehl]

Amelia,

When I had more time today, I compared the display to the current 
version of Amigo and figured out my confusion.  The "63 gene products" 
link is of course tied only to the term, not to the combination of 
qualifier and term.  In the old version, a similar system was used, but 
the link was called "view associations" which seems more directly tied 
to the term itself and not the combination of the qualifier and the 
term.  On the other hand I like seeing immediately the number of gene 
products associated with the term, and as you point out switching the 
order of the columns might make this more clear. The number too, is a 
bit confusing, as it does not seem to correspond to the number of genes 
listed when you click the link, but perhaps I am missing another subtle 
feature here as well.

It is also true that in a logical sense NOT annotations are really 
different from regular annotations and when we make a NOT annotation, we 
are specifically saying that a given gene product is not associated with 
a given function, process, or component.  Thus NOT annotations should 
ideally be excluded from the count of gene products associated with a 
term, just as when people do term enrichment analysis or GO slim 
analysis, we recommend they exclude the NOT annotations.

But it comes down to trying to display the greatest amount of 
information in the least amount of GUI space, and in most cases the new 
version of Amigo succeeds admirably at this task.  I'll stop fussing now 
and see if anyone else cares to comment here,

Thanks,

Alex


Amelia Ireland wrote:
> Hi Alex,
>
>   
>> http://goweb-dev.stanford.edu/cgi-bin/amigo/gp-assoc.cgi?gp=MGI:MGI:2685628&session_id=1458amigo1197932357
>>
>> the first line of the listed associations reads Qualifier: NOT, Term: 
>> GO:0051092 : activation of NF-kappaB transcription factor, link: 63 gene 
>> products,
>>     
> [snip]
>
> I guess this has the potential to be confusing. Currently the gene product
> counts displayed by AmiGO do not include the 'not' annotations, but
> presumably users may still want to know the number of gene products
> associated with a certain term even if for a given gene product, the
> term association has the operator 'not'.
>
> One possibility might be to swap the order of the columns so that the
> qualifier / operator column appears next to the evidence code column
> instead.
>
> Do other people find this confusing, and if so, would swapping the columns
> help?
>
> Thanks,
> Amelia.
>
>   


-- 
Alexander Diehl, Ph.D.
Senior Scientific Curator
Mouse Genome Informatics
The Jackson Laboratory
600 Main Street
Bar Harbor, ME  04609

email:  adiehl at informatics.jax.org
work:  +1 (207) 288-6427
fax:  +1 (207) 288-6131