Making Rational Protein Design “Intelligent”
Richard Chen, Jimmy Kim, Jason Wong, Stewart Slocum
What is Protein Design?
Predicting amino acid sequence of a protein that folds to a given structure.
It's an NP-hard (possibly NP-complete) problem.
Protein Design Holds Promise
Challenge
Number of possible amino acids grow exponentially as chain size increases.
Ways of Computing Sequences
- Optimization problem to minimize free energy
- Discrete problem as for each amino acid spot there are only a limited number of possible rotamers and 20 possibilities
- Current strategies use dead-end elimination then branch and bound for variable selection
X_i=\{r_i, a_i\}
What We Did
Used a DNN to learn strong branching decisions to speed up computation in Toulbar 2.
Features
Trained by taking static, dynamic, and dynamic optimization features at the current execution of the program.
Some examples (of many):
- Average unary cost of domain values in current variable
- Domain size
- Number of nodes solved for
- Weighted degree of current variable (weighted constraint heuristic)