Google DeepMind Adds Cheminformatics Skills to Google Antigravity for Compound Search and Macrocycle Design

In an article dated November 3, 2026, Google DeepMind introduced Cheminformatics Skills for Google Antigravity, covering drug-discovery tasks from literature mining and molecular analysis to proposed designs and synthesis routes. Its example workflow searches thousands of compounds, runs matched molecular pair analysis, and proposes rigidified macrocycles with synthesis routes in an hour. The article says a typical iteration of the described workflow takes multiple days.

The skills are grounded in RDKit and connect to curated resources from ChEMBL and PubChem. Google DeepMind says the language model handles workflow orchestration, database navigation, and scientific reasoning, while deterministic chemistry tools perform molecular calculations. An interactive viewer supports inspection of compounds and metadata and plotting relationships across the dataset. The article also describes benchmarking machine-learning architectures and feature sets, and proposing macrocycles with predicted potency.

The skills are available on Google Antigravity. Individual users can download the app and install the Science plugin; enterprise users can start with “Use Business Account” and Google Cloud credentials. Google is offering up to 10,000 scientists the highest usage tier free for six months, with rolling onboarding and eligibility, capacity, and selection conditions. The article stresses that computational designs remain hypotheses that must be synthesized and experimentally validated.