How To Get Started With Microarray Gene Expression Analysis?

Question

[mtyler.jason]

Hi all,

I have a set of cancer cell lines for 25 patients and I have their corresponding probe set intensities. I wanted to do some analysis like when I used a drug in all the cell lines it killed some the cancer cell lines but not the other. I wanted to know how I can use the probe sets intensities for the 25 patients to see how they behave differently with the drug. Just some analysis why they are reacting in a particular way to the drug while the others aren't. Suggestions how I should proceed?

[Sean Davis]

We can do our best to give you help, but, at the risk of sounding patronizing, the best solution to your problem is to find a local bioinformatics collaborator to work with you on your data.

[Michael Schubert]

Also: try to talk to them BEFORE collecting the data, so they can help you with study design for sufficient statistical power.

[Obi Griffith]

Do you have expression data before and after treatment with the drug? Or do you just have baseline expression data and want to know what differences in baseline expression might explain their differential response to the drug?

[mtyler.jason]

@Obi Griffith. I just have some baseline expression data and I want to know the difference in the baseline expression that can explain the response to the drug. Exactly like you said.

Answer 1

[JC]

Well, if you want to learn how to do it, you can start with R-Bioconductor, in particular Limma: http://www.bioconductor.org/packages/2.11/bioc/html/limma.html

Answer 2

[Irsan]

If you are not comfortable with R nor willing to learn it, purchase a license for an analysis workbench with a graphical user interface like partkek genomics suite or nexus. But I think before you start diving into the data you have to be very clear what your question is, what do you want to compare with what and why?

Answer 3

[Leandro Lima]

I strongly recommend MeV: http://www.tm4.org/mev/