Protein design is a growing field with applications in medicine, green chemistry, and nanotechnology. There is a gap between current computational design methods, which sometimes fail to produce viable initial designs due to insufficient accuracy, and experimental design methods like directed evolution that require an initial level of function for optimization. Alchemical free energy methods, which utilize molecular dynamics simulations to produce more accurate free energy predictions, have the potential to bridge this gap. Multisite λ dynamics (MSλD) is an alchemical free energy method pioneered by Charles Brooks that is uniquely suited to computational protein design because of its scalability to massive sequence spaces. Initial calculations of T4 lysozyme folding free energies demonstrated the accuracy of MSλD through unprecedented agreement with experimental measurements. Subsequent studies of ribonuclease H demonstrated the ability of MSλD to design within spaces of over 30,000 sequences with 15 mutations, while simultaneously giving insight into evolutionary selection pressure.

From folding free energies to protein design with multisite lambda dynamics

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.