Open warrickct opened 5 years ago
I suspect that any trivial scoring system will be not terribly interesting.
phi/psi unfavourable values should arise as a 'physics' consequence of the underlying representation rather than something that has to be added on top. This is somewhat true already - (i.e. values like (0, 0) are excluded when the collisions are on), I suspect it will be more realistic when we add side chains.
Scoring system that follows e.g. number of hydrogen bonds formed when assembling a protein-peptide system or when finding optimal conformations of a single polypeptide. Would need to penalize phi/psi preference violations.