This is a test version of Biostars. For the public version, visit https://www.biostars.org.
GO - micro & macro categories - pvalue

Hi there!

I’m analysing a bulk RNA sequencing dataset (murine data). After having performed the differential expression analysis, I defined the DEGs that I used in order to perform the pathway analysis with Gene Ontology signatures (downloaded from MSigDB, the R package). In this way I got for each pathway a p-value.

The Gene Ontology is divided in three macro-categories, and in particular I was interested in “Biological processes” (BP) that contains more than seven thousands signatures (I performed the analysis with all these signatures).

But I was wondering, is it possible to “merge” these seven thousands signatures in categories? I mean, I would like to have - GO - GO_BP - X categories - and for each category Y GO_signatures. And, moreover, I would like to calculate a p-value for each category.

For instance, if I am interested in “Extracellular matrix” signatures, how can I define a p-value for this category?

All suggestions will be appreciated.

Thanks

Giulia

pathway analysis go pvalue
• 124 views • link
written by GiuliaAC • GiuliaAC

You didn't specify how you were calculating the p-values or your code to perform the analysis, but you can have as many terms as you want in your term database. Most programs use the hypergeometric test (or some similar variant) to calculate a p-value for each term. The p-values are then corrected for multiple comparisons using a method such as Benjamini-Hochberg. It would be similar to using the phyper(...) function in R for each term, followed by p.adjust(..., method="BH") for the p-value calculated for all terms.

• 0 views • link
written by rpolicastro rpolicastro

I performed the analysis with gprofiler2 R package. Sorry, but I didn't understand how can I get ONE p-value starting from X pvalues. Have you any ideas about how to create categories with gene ontologies?

• 0 views • link
written by GiuliaAC • GiuliaAC

0 answers

No answers yet.

Log in to answer this question.

More posts like this