Accelerating the search for antimicrobial peptides with Google Antigravity
deepmind ·
Google DeepMind has introduced Google Antigravity, an agentic development platform designed to accelerate the discovery of antimicrobial peptides (AMPs) in response to rising global antibiotic resistance. The system utilizes ProtNLM, a deep learning model that predicts biological functions from raw amino acid sequences, allowing researchers to annotate millions of previously uncharacterized proteins in the UniProt database. In a recent technical walkthrough, biologists used this tool to identify a novel bacteriocin candidate with predicted activity against pathogens causing respiratory tract infections. The AI agent collaboratively guided the researcher through database filtering, structural validation using AlphaFold models, and genomic context analysis to verify the candidate's novelty and potential mechanism of action. A key feature of the platform is the Biomolecular Structure Viewer, which provides a shared 3D canvas for interactive collaboration between scientists and AI agents. This integrated workflow compresses fragmented scientific processes into single sessions, significantly reducing the time required to prioritize high-confidence leads. The identified peptide candidate is currently undergoing wet-lab validation to confirm its efficacy experimentally.
deepmind