Inductive Bio, an AI drug discovery partner developing virtual chemistry labs, announced that it has joined Anthropic’s life sciences ecosystem and launched a Model Context Protocol (MCP) connector that makes its ADMET (Absorption, Distribution, Metabolism, Excretion, & Toxicity) prediction models accessible directly within Claude. Any scientist can now reason about key drug properties for any chemical structure in natural language directly from Claude’s interface.

The connector gives Claude and Claude Science, Anthropic’s AI workbench for scientists, users on-demand access to predictions for ADMET properties, bringing trusted insights into the same place that scientists are already thinking through key questions about their discovery project. Drug discovery scientists can work with Claude to explore chemical structures, get predictions of ADMET properties, and reason about those predictions alongside the rest of their compound design rationale.

ADMET represents one of the most persistent bottlenecks in small molecule drug discovery: balancing a molecule's potency against a multitude of competing properties that drive how the molecule behaves in the body is essential for a drug’s success yet consumes an enormous amount of time and resources. Inductive Bio's ADMET models are independently validated as state-of-the-art, having placed first among more than 370 entries from large pharma and AI companies in the OpenADMET-ExpansionRx blind challenge. By making these models available to AI assistants, the new connector democratizes access to those capabilities, helping scientists design higher quality drug hypotheses.

“Scientists are increasingly reasoning through their work with Claude, and predictive ADMET models are the perfect complement to add to their toolkit,” said Ben Birnbaum, Inductive’s Co-Founder & CTO. “Any drug discovery scientist can now make use of Inductive’s ADMET predictions in their conversation and immediately reason about what it means for their compound designs and program hypotheses. It’s a step change in access to the state of the art in small molecule drug discovery, and a powerful taste of what our models can do.”