I'm generating UpSetPlots with a large number of interactions (15 groups). There are several hundred interactions within that group. Consequently, when I try to plot that, it's a total mess.
I know how to reduce and increase the number of points shown, but when I reduce it, the plot only shows the lowest level interactions (i.e., overlap only between 2 or 3 groups are shown).
Can I rearrange it so it only shows overlap where, let's say, more than 12 of the 15 groups are present in that interaction?
2 answers
I found the ComplexUpset package very handy when you have multiple interaction plots, you can choose which interaction you want to plot and it is easy to manually arrange your plot : https://krassowski.github.io/complex-upset/index.html
I ended up just brute-forcing it. I summed up the hits of all my interactions, then sorted that column largest to smallest and deleted all hits under a certain threshold.
It's pretty inelegant but it works in a pinch.
If anyone knows of any better alternatives, I'd appreciate the input.
Log in to answer this question.
-
Why number of interactions decreased in current string database for plasmodium
written by rohitsatyam102
-
Extract Genes from ggvenn
written by axe880 •
-
batch correction on circadian correlated RNA-seq time-series
written by sashapokh •
-
Interaction terms and main effects in DESeq2
written by Aspire
-
Protein-Protein interaction modelling without 3D structure
written by michael •
-
Filtering GO terms by neuromuscular processes
written by Swimming bird •
-
GeneMania with single input gene
written by beacska19 •
-
"gene-gene interaction" databases in human
written by Omid
-
ARACNE plugin in Cytoscape
written by pegah •
-
Tutorial: Revisiting the FastQC read duplication report
written by Istvan Albert
