Systematic optimization model and algorithm for binding sequence selection in computational enzyme design.

Area of Science:

• Computational chemistry
• Biocatalysis
• Protein engineering

Background:

• Enzyme catalysis is crucial for biological processes.
• Designing novel enzymes with specific functions remains a challenge.
• Computational methods are increasingly used to guide enzyme design.

Purpose of the Study:

• To develop a systematic optimization model for selecting binding sequences in computational enzyme design.
• To reduce the activation energy barrier by minimizing binding energy between the active site and transition state.
• To evaluate the predictive power of the developed design methodology.

Main Methods:

• Utilized transition state theory of enzyme catalysis and graph-theoretical modeling.
• Represented the reaction system's saddle point using catalytic geometrical constraints.
• Employed a novel heuristic global optimization algorithm for hyperscale combinatorial optimization.
• Performed sequence recapitulation tests on native active sites for two hydrolytic reactions.

Main Results:

• The model successfully identified most native binding sites when considering catalytic geometrical constraints and substrate structural motifs.
• The optimization approach effectively minimized the binding energy to reduce the activation energy barrier.
• The heuristic algorithm proved effective in tackling the complex optimization problem.

Conclusions:

• The developed computational model shows high accuracy in predicting native enzyme active site sequences.
• This methodology holds significant potential for the rational design of novel enzymes with desired catalytic activities.
• Accurate prediction of active site sequences is key to creating enzymes for targeted chemical transformations.