# R: statistical test

I am working on 4 different species of tomatoes. From the data I had, I looked at the occurrence of a particular "event" in certain intervals of their genome (this interval is identical in all 4 plants) and I have a file for each of the species with their probability of occurrence. The file looks something like this:

ch01:a1-b1
p = 0.45
ch01:a2-b2
p = 0.005
...
...

I have 4 such files (one for each species) where the chromosomal locations/intervals are identical and the probability value differs. Now I would like to find out those intervals where the occurrence of this event stands out.

What do I mean by that? If I find, for example, over a particular interval, the event is more significant in plant/file 1 and 3, but not the others, then this interval is of interest to me. I would like to find all such intervals where "the occurrence of event is not all equally significant/insignificant".

My question is how should I go about this? Which parametric/non-parametric test can I use? Can I use the test on each interval separately? What are the usual issues to look for and overcome.

-