Google reports faster protein design and adaptive robotic lab experiments

Google Research reports that its protein-design pipeline accelerated computational cycles more than eightfold by limiting a costly folding calculation to the local binding pocket. In laboratory experiments targeting fibroblast activation protein alpha, a receptor overexpressed in solid tumors, its top designs achieved single-digit nanomolar binding affinity. In separate experiments, AI agents used laboratory results to revise DNA sequences that bind proteins, reaching confirmed on-target hit rates of 33.5% for myoglobin and 42.1% for α-thrombin after three rounds. Google also demonstrated a digital laboratory model that helped agents reject physically invalid robot commands and recover from equipment failure. Combining that automation with closed-loop protein discovery is its next step.