This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Enrichment Analysis

What I have done so far: (mouse samples)

  1. Used salmon for mapping, fishpond for differential gene analysis
  2. Used fgsea for GSEA with hallmark gene sets downloaded from MSigdb
  3. Used fgsea again for GO term enrichment analysis with GO gene sets downloaded from MSigdb (all genes from fishpond were used, ranked the gene list with signed fold change multiplied by -log10(p-value), no filtering of any kind)
  4. Used goseq for GO term enrichment analysis (filters the genes with p values < 0.05, used the average of the length of each gene from salmon for all samples for bias.data in nullp())

Question: For GO term enrichment analysis, fgsea and goseq generated totally different top hits. I wonder which results I should trust more. What are your opinions?

rna-seq fgsea goseq
• 1,103 views • link
updated by Istvan Albert Istvan Albert • written by scsc185 • scsc185

1 answer

Some things never change, check out this 7 year old thread, the explanations still apply

Why does each GO enrichment method give different results?

I'm new to GO terms. In the beginning it was fun, as long as I stuck to one algorithm. But then I found that there are many out there, each with its own advantages and caveats (the quality of graphic representation, for instance).

and also:

  • Interpretation of biological experiments changes with evolution of the Gene Ontology and its annotations

https://www.nature.com/articles/s41598-018-23395-2

• 0 views • link
written by Istvan Albert Istvan Albert

Log in to answer this question.

More posts like this