Protein design has undergone a revolution in recent years, driven by advances in computational methods and machine learning. The field encompasses two broad approaches: redesigning existing proteins to improve or alter their properties, and de novo design of entirely new proteins with structures and functions not found in nature. Both approaches seek to solve the inverse protein folding problem, determining which amino acid sequence will fold into a desired three-dimensional structure with targeted functional properties.
David Baker's laboratory at the University of Washington, using the Rosetta software suite, pioneered computational protein design and demonstrated the creation of novel protein structures including barrels, cages, and assemblies. Baker co-founded the Institute for Protein Design and companies like Neoleukin Therapeutics, which designs novel cytokine mimetics for immunotherapy. The field accelerated dramatically with the advent of deep learning approaches, as companies like EvolutionaryScale (developers of ESM protein language models) and Generate Biomedicines brought AI-driven protein design to bear on therapeutic and industrial problems at unprecedented speed and scale.
The commercial applications of protein design span pharmaceuticals, industrial enzymes, biosensors, and materials science. Designed proteins can serve as therapeutic candidates with optimized binding affinity, stability, and immunogenicity profiles. In industrial biotechnology, designed enzymes can catalyze reactions under conditions that would denature natural enzymes. The convergence of protein design with high-throughput experimental validation, including display technologies and next-generation sequencing, is creating rapid feedback loops that continuously improve the accuracy and ambition of computational design methods.