Agents and CodeHigh importance

DeepMind launches Cheminformatics Skills in Google Antigravity for agentic drug discovery

Google DeepMind announced Cheminformatics Skills for Google Antigravity: the model orchestrates literature and database search, analysis, and candidate proposals, while deterministic tools (RDKit and ChEMBL/PubChem connectors) run the molecular calculations. In a demo with lung-cancer treatments (crizotinib/lorlatinib), the agent retrieves thousands of compounds, runs matched molecular pair analysis, trains potency models, and proposes rigidified macrocycles with synthetic routes in about an hour — a workflow that previously took multiple days. The company stresses that computational designs are hypotheses to validate in the lab. Skills ship today in Antigravity’s Science plugin; there is a six-month offer at the highest tier for 10,000 eligible scientists.