GEO analysis with R

Question

[abdalla.m]

Hi,

I have started using R extensively to run some scripts and analyse data. In the past few weeks, I have been using GEO query to download datasets and analyse it.

My question is:

How do I know if what I am doing is correct? I can handle the errors when it arises, the plots are always nice, the genes I get either differentially expressed or not, make sense. But, I always doubt that I may be mistaken?

Any comments?

Thanks,
Abdalla

[seidel]

If it's a GEO query, then often there will be a paper associated with the data and the results derived from them. Have you checked to see if you get the same results as those in the publications you may be re-analyzing?

the genes I get either differentially expressed or not, make sense.

This implies you've tried the suggestion above, and is a good sign. Another thing you can do is to make sure the data is internally consistent. Check the individual data points for genes with low p-values to see that they resemble differential expression vs other genes in the set with higher p-values, etc.