Determine statistically significant differences in protein sequences

Question

[brett.spurrier]

I have a list of 100 homologous protein sequences wit various differences between them.

I also know that a percentage of them bind to a target (control group). So I am trying to learn a systematic way to identify the differences in amino acids between the control group and non-control group. Preferably wit some statistical confidence. Obviously, a sequences alignment would "show" me differences, but I'm trying to identify the important ones.

Can someone point me to some keywords or applications that I could learn about to do this?

[Asaf]

I'm not aware of a software that does specifically that but it's a simple Fisher's exact test or chi-square, testing if each sequence "feature" is enriched in the non-control. Although you'll have to take evolution into account, if your trait is evolutionary conserved and so are the sequences you will probably find correlating changes which are unrelated.