Statistics problems with genetic analysis

#1
hi everyone,

I'm having a bit of trouble trying to find out what to do with some results I've got in my research project. I've found 2 different groups of DNA sequences which I would expect to be different on their composition, so I got the results for the composition in both groups, the sample number is small (< 50) but apparently they don't follow a normal distribution, so I couldn't use an ordinary t-test using R. What I'd like to find is the statistical validation to confirm that these 2 groups either differ or are statistically the same.

Thank you very much for your help!

rodpck
 
#2
What is it that you have measured in this experiment? Is it some sort of sequence allignment? Some sort of score of similarity? If this measure is simply a quantitative score that identifies the composition of the sequence, then perhaps you could use a non-parametric procedure to test for a difference in the medians of the two groups. These types of tests are distribution-independent, and therefore do not require normality. If these data are a set of sequence allignments, other, more powerful (and complex) procedures have been developed to manange these situations.

~Matt