It is going to take more than a couple of algorithms to solve this problem.
Top biocomputing scientists have been working on this problem for years and
years, with hundreds of thousands of pounds worth of grants to fuel them.
Have they come up with prediction software that is anywhere near useful?
Good place to start finding more about it is the Crystallography and
Biocomputing www page on the Cambridge University Biochemistry server
go from there and be bewildered!