Microsoft Research’s Quine accelerates drug discovery with integrated AI model

Microsoft Research introduces Quine, an AI system integrating genomics, chemistry, and bioimaging to speed up early-stage drug development, showing promising results in pancreatic cancer research and aiming to reshape laboratory prioritisation. Human review remains essential as the tool remains in experimental phase.

Microsoft Research has launched Quine, an experimental artificial-intelligence system aimed at speeding drug discovery by linking computational biology with laboratory work. The project matters because drug development is still slowed by expensive, time-consuming screening, and Microsoft is betting that a broader biological model will help scientists focus resources on the most promising compounds. If it works, the approach could shorten early-stage research and reshape how life-sciences teams prioritise experiments.

According to Microsoft, Quine combines a unified “world model” spanning genomics, proteins, chemistry, cell states and bioimaging, rather than treating each discipline as a separate problem. It was built with the Broad Institute of Harvard University and MIT, and is designed to help researchers move from a question to hypotheses, predictions and experiments.

The company says early tests in pancreatic ductal adenocarcinoma cell lines allowed the system to rank compounds that produced a switch in cell state within a single weekend, while also surfacing unexpected phenotypic responses. Microsoft stressed that Quine remains a research tool, not a clinical product, and that human review is essential because AI systems can err in analysis and inference.