you can do test of significance by  student t test 

I think it depends on what aspects of changes you are trying to look at.

You need to compare the ratios of different types of mutations so Fisher's test is better.  There is a very old software for comparing different mutational spectra and can give you a p value.  I don't know if it still work or not.  I used it a lot many years ago.  You may try it.

Computer program for the analysis of mutational spectra: application to p53 mutations.

Neal F. Cariello, Walter W. Piegorsch, W.Thomas Adams and Thomas R. Skopek. Carcinogenesis (1994) 15 (10): 2281-2285.

Hi, you can find an introduction to the topic in this article from the Ostermeier Lab:

http://ostermeierlab.johnshopkins.edu/pdf/21_BiomolEng.pdf

I hope it helps!

Fisher's test and t-tests will be useful to you. There is one paper of Cariello on the analysis of mutational spectra . It was a paper of 1994 or 1995. I am not remembering the journal but you can google search it. In addition following papers may be useful.

1) ESSENTIALS: software for rapid analysis of high … - ‎Zomer - Cited by 53

2)  Statistical test for the comparison of samples from … - ‎Adams - Cited by 250

I hope this information may be useful to you.

Many thanks!! you all are very kind!

What about statistical analysis to check if you are really sampling the diversity conviently? How do you know if you are taking a representative sample to measure diversity?

maybe rarefaction curves?

Best

you should check this server and its publication, which is specific for epPCR libraries...

http://mutanalyst.com/

And in this paper from one collaborator you will find other useful resources:

http://stat.haifa.ac.il/~yuval/papers/2014-MIMB-prob-methods.pdf