[R] Statistics newbie question

[James Noble]

Hi there,

I am relatively new to R and statistics.

I have large dataset where each sample has about 3000 dimensions, 
and belongs to one of 13 classes. I am wanting to measure the statistical
significance of each dimension with respect to the class, in order to
find what dimensions are likely to be good predictors.

I undertand the standard way to do this is with ANOVA or regression, 
but am not to sure if these methods are the most suitable. 

If this is a good way to approach it how do I code it as  a regression
problem in R.

Thanks for your help.

-James