Using Enrichment Map and Autoannotate for Gene Enrichment Networks

[Sayantan Laha]

Hi,

    I am trying to generate figures in the manner shown in the attached images. The above are from the paper titled "Epigenetic analysis of Parkinson's disease neurons identifies TET2 loss as neuroprotective" by Marshall et. al.

I am working with a set of differentially expressed genes. I have used gprofiler to generate the gene enrichment map file with my gene set. Then I have used EnrichmentMap module of Cytoscape to generate the network and Autoannotate to generate the clusters. A subset of my genes are regulated by miRNA and a subset by methylation, with an overlap between the two sets. I am looking to color my clusters according to the mode of regulation. For example, if I assign clusters "regulated by methylation" as red and those "regulated by miRNA" as green, then those regulated by methylation and miRNA should be a mixture of red and green, in the manner mentioned in the aforementioned paper. Using Autoannotate, I am unable to assign color to the clusters manually, as the color I select gets applied to all the clusters. Moreover the palette option assigns colors to all clusters automatically, without specifying the method used to determine the color for each cluster.

Is this possible using just autoannotate or do I need to download anything else?

Thanks,

Sayantan

[Scooter Morris]

Hi Sayantan,

I'm a bit confused.  You can certainly use the bypass (third column) in the Styles panel to set whatever color you want to nodes, including the background cluster color (which are actually also nodes).  This will override whatever color AutoAnnotate initially chose.  My confusion is whether you want to color the cluster color (i.e. the background oval) or individual nodes.  Obviously, coloring individual nodes using a bypass would be onerous, so you would want to set up (or change) a mapping for the fill color using a discrete mapper (assuming you have a column that tells you if the nodes are regulated by methylation or regulated by miRNA).  Then, you would color the cluster (background oval) manually using a bypass.  Does that help?  If not, it would be useful to understand a bit more exactly what you are trying to do.

-- scooter

[Ruth Isserlin]

Hi Sayantan, 

Through the AutoAnnotate panel changing the colour of the annotations changes all of them but the annotations are also listed in the separate annotation panel on left hand side.  You can select the annotations (one at a time unfortunately) to change the setting like colour/shape/opacity ....

The fancy half purple and orange annotations in the referenced paper were not done in cytoscape, I don’t think.  For effects like that I think they exported the image and modified it in illustrator or another editor.  

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 view this discussion on the web visit https://groups.google.com/d/msgid/cytoscape-helpdesk/e7b90993-fe50-446e-aff0-8a15ef2f6418n%40googlegroups.com.