[BioC] Reporting results of HyperGTest

[Pankaj Chopra]

Might be worth trying 'GOGraph' in the GOstats package. I think I was 
able to do it using it with the Rgraphviz (?makeNodeAttrs) package.




Paolo Innocenti wrote:
> Hi all,
>
> I performed two separate HyperGTest (pkg:GOstats) on different list of 
> DE genes (trt1 vs control and trt2 vs control). The results are good and 
> in the direction I was expecting: I can see from "htmlReport" that the 
> GOterms overrepresented in each list support my hypothesis.
>
> My problem now is to "quantify" and to "show" what I can see: I am 
> looking for a nice way to format my results. I have around 50 GO nodes 
> in each list, too many to list them all in two colums. Is there a 
> graphical way (or a compressed table) to show differences in "which 
> branches of the GO" are overrepresented, and possible overlaps?
>
> I am basically looking for something in between htmlReport, a DAG graph 
> and the function ontoCompare in goTools.
>
> I appreciate any comments on this issue. I have no experience with DAG 
> graph, so if you suggests that it is the way to go, I'd like to be 
> pointed to some documentation (I found GOvis.pdf too confusing as a 
> beginner)
>
> Many thanks,
> Paolo
>
>
>