[BioC] Running the GSEA algorithm

[Julian Gehring]

Hi Enrico,

You can also have a look at the GSRI package:
http://bioconductor.org/packages/release/bioc/html/GSRI.html

Best wishes
Julian


On 31/03/14 21:19, Enrico Ferrero wrote:
> Hi everyone,
> 
> I would like to include GSEA into my R analytical pipelines, but I'm
> struggling to understand what's the best way to implement it. The following
> information might be incomplete or even wrong, but here is what I
> understood so far:
> 
> - The GSEABase package [1] provides an excellent infrastructure for dealing
> with gene sets and gene sets collections but, as far as I understand,
> doesn't provide a way to run the GSEA algorithm.
> 
> - The PGSEA package [2] provides a minimal, and perhaps simplistic,
> interface to GSEA. It does run the analysis but only outputs a matrix with
> what I understand is a score (possibly the NES?) and nothing else.
> 
> - The SeqGSEA package [3] allows to run the GSEA algorithm and also
> produces some excellent plots of gene sets enrichment. However, it works
> with with RNA-seq count data and I don't see how it could be adapted to
> microarray data.
> 
> - The GSEA-P-R package from the Broad Institute [4] is arguably not ideal
> to work with and its use is basically discouraged.
> 
> - The Java version of GSEA [4] is probably my best bet at the moment, as it
> allows command-line usage and provides a complete output for the analysis.
> 
> So, am I missing something here?
> Is there an established way to run the GSEA algorithm from R using
> Bioconductor packages that also works for non-NGS data?
> If not, would anybody recommend the GSEA-P-R package from the Broad
> Institute?
> Are there any other options?
> 
> Thanks very much.
> Best,
> 
> [1] http://www.bioconductor.org/packages/release/bioc/html/GSEABase.html
> [2] http://www.bioconductor.org/packages/release/bioc/html/PGSEA.html
> [3] http://www.bioconductor.org/packages/release/bioc/html/SeqGSEA.html
> [4] http://www.broadinstitute.org/gsea/downloads.jsp
>