Agentic cheminformatics with Google Antigravity

Google DeepMind built a suite of Cheminformatics Skills for Google Antigravity that it says accelerates drug discovery workflows, shrinking an iteration that usually takes days to about an hour.

• The skills turn Antigravity into a collaborative science workbench with local tooling, aimed at chemists who spend too much time wrangling molecule data instead of testing hypotheses.
• In a walkthrough on rigidification, an agent retrieved thousands of compounds from lung cancer literature within minutes, filtered out buffers and solvents, and documented their sources in a report.
• The underlying language model handles orchestration and scientific reasoning while deterministic tools, grounded in the open source RDKit and connectors to ChEMBL and PubChem, run molecular calculations.
• The retrieved set was explored in an interactive viewer where scientists can inspect molecules and plot potency against rotatable bonds, showing that the most rigid compounds are the most potent.