I wish to calculate the energy for a protein where the position of all the C-alpha atoms is known.
One way is to calculate the pairwise distances between the atoms and then look up a probability distribution to determine the likelihood that the distance for that particular pair is the given value, take the negative logarithm of that value, sum this over all pairs and this is a proxy for the molecular energy. This is the approach taken by the machine learning community. (Physics-Based Design of Protein-Ligand Binding)
Can anyone point me to a sample calculation or publication where this is elaborated in some detail?
I have very little background in chemistry beyond the introductory college courses. The question that I am trying to research is this: Given two spatial configurations of a molecule, how does one determine that one configuration has a lower energy than the other? What is the benchmark that is used for measurement? When Ramachandran generated the plots of torsion angles, how were the optimal angles calculated?