[go] AmiGO beta announcement

Michael Ashburner ma11 at gen.cam.ac.uk
Tue Dec 18 02:06:41 PST 2007



Seth and the Amigo working group

This looks great, and so fast !

I agree with Val that we should really encourage people to use full  
id's, ie. GO:0000123, and not 0000123 or 123.

I guess the Help page has not yet been redone, as this directs to the  
old Help.

I have not tested the GO Slimmer (I don't mind the name as much as  
some !) but does it allow
the user to keep his slimmed ontology (ie without any instance data) ?

When you look at  a graphical view why is "all" repeated in the root  
box ? Does this come from:

[Term]
id: all
name: all
namespace: universal
def: "This term is the most general term possible" []


Why are some, but not all, GO slims hyperlinked ? eg from the page  
for "metallochaperone activity"
there are two subsets (slims?, why use two words?), and only the pir  
one is linked:

         * goslim_pir
         * gosubset_prok

(ah, is it just gosubset_prok that is never linked ?)

I see treeview is limited to 500 lines. Is this for the server's  
benefit or the user's ? If the latter could there be a "view all"
option ?

Looking at the PIR slim in Treeview I got this when invited to try  
"compact view of the graph":
     *  fatal message   AmiGO could not find any terms to perform  
your query upon. Please go back and select one or more terms.

and this when invited to see the old GO format:

$all ; all
  %biological_process ; GO:0008150
   %reproduction ; GO:0000003

Indeed the same terms in OBO or RDF format. Something wrong here.


Thanks all for the magnificant work.

Michael
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://fafner.stanford.edu/pipermail/go/attachments/20071218/5c1ac33b/attachment.html 


More information about the Go mailing list