In CytoScape, how to efficiently group/cluster gene sets from the enrichmap according to their functionality (pathway-wise)

[weiqi.w...@gmail.com]

 

Dear sir or madam:

Hi I just started using CytoScape and found its really a great tool! Long story short, after all those powerful processing that CytoScape enables, I went to one of the final stages where I want to group the genesets from the enrichment map according to their functionality. The attached pictures should be of a better clarification of what I wanna do.

To be more specific, after all the previous processing steps I got to see the enrichment view, where all the genesets were clustered according to an ambiguous "similarity" measurement, so that highly similar gene-sets are placed close together. This is my Q1: what exactly is this "similarity" measurement? What are the detailed criteria?

When I was wondering what to do next, I found a tutorial with a figure of grouped gene sets according to their clearly-labeled functionality, and this is what I wanna do. I noticed that the Figure legend clearly says "The network was manually rearranged to improve layout, and major clusters were manually labeled", therefore I was keen to know how exactly this was done -  this is my Q2: is there an automated way to group the genesets? e.g. group all the genesets that majorly concern rRNA processing, or cell cycle, etc.? Does CytoScape have such a function?

Thanks so much for your help in advance, I look forward to your reply!

yours

weiqi

[Ruth Isserlin]

Hi Weiqi,

>Q1: what exactly is this "similarity" measurement? What are the detailed criteria?

It depends a little on your dataset and what parameters you chose when you built your enrichment map.  The similarity measure can be either calculated using  jaccard, overlap or combined coeefiecient. (the formulas are outlined here http://www.baderlab.org/Software/EnrichmentMap/UserManual#Jaccard_Coefficient ) and I have attached a small picture of what tuning this paramter does.

​

One other variable is whether or not you have used an expression set.  If you have used an expression set then the genesets in your network are restricted to contain only the gene in the expression set and the union and intersections calculated are based on this smaller set.  If you haven't used an expression set then genesets are the sizes indicated in geneset file.  

>Q2: is there an automated way to group the genesets? e.g. group all the genesets that majorly >concern rRNA processing, or cell cycle, etc.? Does CytoScape have such a function?

Unfortunately this is still a manual process.  We currently have a prototype of the automated annotation feature but it only works in cytoscape 3.2.  It should be released very soon though. 

Thanks, 

Ruth  

--
You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cytoscape-helpd...@googlegroups.com.
To post to this group, send email to cytoscape...@googlegroups.com.
Visit this group at http://groups.google.com/group/cytoscape-helpdesk.
For more options, visit https://groups.google.com/d/optout.