GSEA with variant proportions as input

Question

[manali.rupji]

Hello All,

I am interested in using using variant proportions as input to perform Gene Set Enrichment Analysis (GSEA). Is anyone aware of a Bioconductor package in R or any other tool that can help me do GSEA analysis using variant proportions data as input rather than the typical RNA seq expression data. I have two different phenotype groups which I want to compare and I can have the proportion data transformed if needed. I also intend to combine these results with RNA seq expression data results.

Doing some research, I have come across many papers using p values but I am mainly interested in using variant data input.

Any help would be greatly appreciated. Thanks,

M

[GouthamAtla]

What kind of data you have ? Tell more about the experiment. Can you give some examples like whats there in rows and columns ? You could try GSEA Pre-Ranked Analysis if you can rank your data somehow. But if you can give more information, people could help.

[Devon Ryan]

In addition to the comment from Goutham Atla, you'll need to carefully consider how you normalize for any apparent GC composition bias or other large coverage differences between samples (since these might affect the variant calling power and lead to aberrant proportion measures). Some of the functions from deepTools might come in handy here (not that I want to intentionally plug some tools from a colleague).