If you use bioconductor as suggested by Juan, there is an easy way to test specific gene lists for differential expression using a function called geneSetTest(). It's a function in the limma library in R, also called "wilcoxGST", which computes a p-value to test the hypothesis that a selected set of genes tends to be more highly ranked in terms of a given test statistic compared to a randomly selected set of genes. (refs below)

Goeman, JJ, and Buhlmann P (2007). Analyzing gene expression data in terms of gene sets: methodological issues. Bioinformatics 23, 980-987.

Michaud, J, Simpsonet al. (2008). Integrative analysis of RUNX1 downstream pathways and target genes. BMC Genomics 9, 363. http://www.biomedcentral.com/1471-2164/9/363

You might also check out GSEA (which can be performed in R).

Take a look at:

http://www.bioconductor.org/packages/release/bioc/html/globaltest.html

Also, just in case you are not aware, you can pull your GEO data into R/bioconductor using:

http://www.bioconductor.org/packages/release/bioc/html/GEOquery.html

Search in the bioconductor.org site for tutorials on the affy, lima or marray packages, they should give you a good start. what kind of data do you have (e.g. file format)?

As I am not bioinformatician could you suggest any other software/website than R/Bioconductor? It is hard to start form scratch with that software.

If you don't want to start with BioConductor, and want something with a GUI you could try MeV and/or TM4 or GeneSpring - the latter is commercial and requires a license.