GAGE or GSEABase using expression data and pathway.gmt file for mouse

Question

[snishtala03]

Hello,

I have mouse data for 8 samples (2 conditions, 4 samples each). I performed RNA-Seq analysis and have expression data for different genes across these 8 samples. I also downloaded pathway file from http://download.baderlab.org/EM_Genesets/April_01_2017/Mouse/symbol/Mouse_AllPathways_April_01_2017_symbol.gmt

I am trying to do gene set enrichment analysis using R packages GSEABase or GAGE but the tutorial is not being very helpful for this case. moreover, since it is mouse data, I am not sure if I can use the R package to do this.

Please let me know your suggestions.

Thanks!

[badribio]

Why would you not use gsea?

[snishtala03]

I need to do gene set enrichment for multiple (20) data sets. I was looking for a R option, so I can automate it and it would be faster.

Answer 1

[jeff.wong]

The Bader group is developing a 'Guide' to pathways and analysis.

An early beta version includes a step-by-step workflows for gene set enrichment analysis (GSEA) using RNA-Seq (http://pathwaycommons.github.io/guide/workflows/pathway_enrichment_ii/index/ ). The workflow itself provides example data; it also goes a step beyond to visualize the results of GSEA in the form of an 'Enrichment Map' (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0013984 ) - a graphical view of the gene sets. It uses the GSEA desktop and Cytoscape software rather than R.

Please take a look. I would be particularly interested in hearing your thoughts on clarifying and improving the workflow.

[badribio]

If you plan to go this route that @jeff.wong suggested, Use bader groups own curated list of gene sets, I have used it in the past and it works fine.

[Lluís R.]

In my screen I see the letters of the workflow in a light grey which makes hard to read the workflow.

Answer 2

[bigmawen]

Gage package has a tutorial on data preparation, which describes how to read and prepare gene set data in gmt format: http://bioconductor.org/packages/release/bioc/vignettes/gage/inst/doc/dataPrep.pdf

GAGE is a gene set/pathway analysis method, which can be applied to RNA-seq data independent of species. You need to make sure you have the gene set data and gene expression data using the same gene ID types (Entrez Gene, symbol, etc). you may generate KEGG or GO gene set data using functions kegg.gset() and go.gsets() for thousands of species in gage package. For details, check the main tutorial of gage package:
http://bioconductor.org/packages/release/bioc/html/gage.html

If you prefer a GUI tool, you may also do pathway analysis, visualization and data integration using Pathview Web server:
https://pathview.uncc.edu/