How can i do enrichment analysis more precisely for differential expression analysis results ?

Question

[Angela]

I did differential expression analysis and I want to identify most related pathways to my leukemia sample, I tried many different websites (toppfun, enrichR, GSEA,) but the results are huge How can I do enrichment analysis more specifically? to fine more related results

[Angela]

Hello b.nota, I got error in installation? had non-zero exit status how i can use gogadget package in R?

[GenoMax]

Please use ADD COMMENT/ADD REPLY when responding to existing posts to keep threads logically organized. Your comment should have gone under @b.nota's answer.

[Benn]

What version of R do you use (what OS, Windows?). Do you have the other required packages installed?

[Angela]

I use R version 3.5 on windows. and installed dependency packages 'gplots'

[Benn]

I am only guessing here (never had installation problems before), did you install all dependencies? So goseq and dplyr as well?

[Angela]

Thank you for your good package

Answer 1

[liux.bio]

You can try Cytoscape APP Bingo. It presents the hirarchical relationships of the GO terms. I often use it to find specific enriched GO terms.

Additionally, you can try Revigo. It summarize GO terms by removing redundant terms. It has awesome visualization.

If you want to explore enriched Kegg pathways, you can try KeggExp. PS: I am the author of KeggExp.

Answer 2

[Benn]

You can try to filter the results, to filter out too general terms (term like "biological process", or "metabolism"). Or/and you can cluster similar GO terms (based on overlapping genes). I made a tool in R with functions to filter or cluster, called gogadget. You'll have to do the analysis with goseq first, or you'll have to convert your results to goseq results data.frame.