Two big challenges in computational protein design: 1. SEARCH PROBLEM: There are many possible sequences: 20N in general, these can t be enumerated exhaustively 2. ENERGY PROBLEM: To evaluate AGfold for a sequence we need to know the energy in the folded and unfolded states a what is the structure of the folded state? we know the backbone but what about the side chain atoms? b, how should we mode the unfolded state? C. it is hard to model the free energy, g d. we need a fast way to evaluate the energies because there are so many sequences to considerTwo big challenges in computational protein design: 1. SEARCH PROBLEM: There are many possible sequences: 20N in general, these can’t be enumerated exhaustively 2. ENERGY PROBLEM: To evaluate ∆ Gfold for a sequence we need to know the energy in the folded and unfolded states a. what is the structure of the folded state? we know the backbone, but what about the side chain atoms? b. how should we model the unfolded state? c. it is hard to model the free energy, G d. we need a fast way to evaluate the energies because there are so many sequences to consider