Quantum-Hybrid Drug Discovery
We combine quantum-enhanced generative models with large-scale virtual screening to identify viable small molecules faster and at a fraction of the cost of traditional discovery workflows.
What are Quantum-Hybrid Approaches?
Generation → Screening → Refinement → Prioritization
In drug discovery, quantum-informed models can be used for molecular generation and search efficiency:
-
Hybrid quantum–classical generative modeling
-
Ultra-large library screening (10⁸–10⁹ compounds)
-
Physics-based refinement (docking, MD, rescoring)
Image by Thomas Splettstoesser (www.scistyle.com)
Case Study: KRAS Inhbitors
Recent work, involving members of Virtual Discovery, published in Nature Biotechnology demonstrates what this looks like in practice.
A hybrid quantum–classical workflow was used to design small molecules targeting KRAS:
-
Training set expanded to 1M+ compounds using VirtualFlow 2.0, STONED, and Chemistry42
-
Screening performed against ultra-large chemical space (including Enamine REAL)
-
15 candidate molecules generated
-
2 compounds showed strong development potential