Protein engineering modifies the amino acid sequences of proteins to improve existing properties or create entirely new functions. Proteins are biology's workhorses — they catalyze chemical reactions (enzymes), provide structural support, transport molecules, and recognize specific targets (antibodies). Engineering proteins with enhanced stability, activity, specificity, or novel capabilities has applications across medicine, industry, agriculture, and materials science.
Two complementary approaches dominate the field. Rational design uses knowledge of protein structure and function to make targeted amino acid changes — for example, modifying an enzyme's active site to accept a new substrate. Directed evolution, by contrast, generates millions of random variants and lets selection identify the best performers. The 2024 Nobel Prize in Chemistry recognized both approaches, honoring David Baker for computational protein design and Demis Hassabis and John Jumper for AlphaFold's protein structure predictions.
AI is revolutionizing protein engineering. DeepMind's AlphaFold can predict protein structures with atomic accuracy, providing the structural knowledge that rational design requires. Generative AI models from companies like Generate Biomedicines and Profluent Bio can design novel proteins from scratch, guided by desired functional specifications. These computational tools are converging with high-throughput experimental platforms to create a virtuous cycle: AI designs proteins, automated labs test them, and the results train better AI models. The pace of protein engineering is accelerating dramatically as a result. For deeper coverage, see SynBioIntel.