Enrichment Score Calculation In Bioconductor

Question

[Layla]

Hi: I have a set of gene signatures obtained by SAM that contains data like this:

Gene ID     Score (sd)   
123456_at      45,5655
.......              ........

where Score is the t-statistics obtained.

The question that I have is based on that data, how I can obtain the enrichment score using bioconductor?

[miquelduranfrigola]

Hi Marco Aurelio. Enrichment score of what? Do you mean GO term enrichment analysis? KEGG pathway enrichment analysis? Do you just want the p-value of your score? Try to be a bit more specific, and it will be easier to help you.

[Layla]

I want to compare this list of genes with a data matrix that contains a set of probe sets ordered by their differential expression between healthy and disease samples.

[Layla]

If I have the set of up and down regulated genes as a signature from SAM, how can I compute an enrichment score for these genes, if I have only the t-statistics? thanks

Answer 1

[seidel]

You could try the geneSetTest() function in the limma library (available at bioconductor). Given that you have gene signatures (i.e. a group of genes), and a healthy/disease data set, you can test your gene signature for significance in a data set as follows:

library(limma)
geneSetTest(gene_sig, dataset)

where gene_sig is a vector of indices representing the genes in your gene group that match genes in your data set, and dataset is your healthy/disease data ordered by some statistic (i.e. ratio or p-value).