Try EnrichR (http://amp.pharm.mssm.edu/Enrichr/) and www.bioprofiling.de, two sites where you can analyze gene lists. To be more specific I would need to know what you scientific project is about.

EnrichR is similar to DAVID and GSEA but it is more complete and extremely easy to use.

Bioprofiling has some additional features.

You might laso wish to take a look at http://hefalmp.princeton.edu/ where you can get information on virtual interactions of your gene(s).

Good luck!

Hello Neus,

GSEA is powerful tool, however depending on your quest,  probably it is worth to perform network analysis. it will give you an idea about the interaction between different targets. One of the tools is EGAN (http://akt.ucsf.edu/EGAN/), but of course there are others. 

Greetings 

Hi Houssein,

Thank you very much for your answer. I will try it, it seems a really neat approach and something worth doing given my data set!

Greetings

Try EnrichR (http://amp.pharm.mssm.edu/Enrichr/) and www.bioprofiling.de, two sites where you can analyze gene lists. To be more specific I would need to know what you scientific project is about.

EnrichR is similar to DAVID and GSEA but it is more complete and extremely easy to use.

Bioprofiling has some additional features.

You might laso wish to take a look at http://hefalmp.princeton.edu/ where you can get information on virtual interactions of your gene(s).

Good luck!

Than you very much for your answer Ulrich!

Yes, sorry, maybe I should have specified I work with Drosophila, though it's a model organism quite common so usually is an option on these programs. 

We up/down activate a signalling pathway in order to find possible targets activated/inhibited by it. The lists I've been working on is the result of comparing each condition with the control (wild type situation), and although I have a nice list and some nice candidates with a visible phenotype, I feel like I could work a little bit more on the results (raw and my final list) in order to be able to get more information out of the microarray. 

I will try all your suggestions, thank you very much.

Greetings!

Depending on what biological response you are looking at, what you can discover using GSEA will primarily depend on your data quality and sample size for the phenotype classes. If you only have n=3 for control and treatment, then unless your treatment elicits large biological response, GSEA will likely fail (as with differential gene expression analysis). My personal experience with GSEA is with using it to mine public microarray data sets where the sample sizes are respectable. You really need that because biological processes are inherently noisy (not to mention the background noise!).

Hello Tsung! 

Thank you very much for answering. Yes, my n=3 for both control and treatment, though every "n" is a sample consisting of a pool of tissue from 4-6 individuals. But I get your point, I'll look into it and try to focus more my attention to other types of tests then, though I'll try to see if I find something relevant with the GSEA.

Thank you very much,

Greetings!

Thank you very much for your answer Thomas. I know it is maybe not the best design but the tissue we were dissecting from was really limited (around 40000cells) and although previous to the Microarray we had RNA amplification, we needed a minimum amount of tissue and that's why the samples are pools. So, in your opinion, given these limitations, we would still be better off if we apply network based methods rather than single gen such as GSEA? I have also thought about PCA, though I am not sure something relevant or interesting would come out of it. Maybe I should just accept the data are not the best for further analysis nor a global trend result and go with the individual phenotypes I get from the phenotypic screening of the targets?

ssGSEA: http://www.broadinstitute.org/cancer/software/genepattern/modules/docs/ssGSEAProjection/4

ssGSEA GP Module: 

http://genepattern.broadinstitute.org/gp/pages/index.jsf?lsid=urn:lsid:broad.mit.edu:cancer.software.genepattern.module.analysis:00270:4

by "bags of genes" I mean a group of genes that together describe a pathway.

Thomas thanks and OK, I'll try to do it and look into PCA deeper.

Alain thank you very much for your answer. I think I can get the Drosophila gene set relatively easy, I'll have to look into it to be sure. Thanks very much for ssGSEA input, I'll try to understand what it is and apply it.

Depending on the specific question you are addressing you might wish to download somewhat related datasets. After pre-processing them together with your samples using RMA default settings (quantile normalization reduces batch effects) you can perform cluster analysis to detect gene sets that distinguish your cells from the other cells or statistical analyses such as multiclass SAM to identify genes that are specific for your cells.

Hi Ulrich, thank you very much for your answer. I will have to look into it because I am not sure how similar other datasets can be. I know microarrays have been done with the same pathway, but the conditions are not the same, so I don't know how much comparable that is, but thank you very much for the idea, I will definitely try it at least. I am so grateful because a lot of things can be done that I had never heard about.