QC Ware and IonQ have demonstrated a hybrid quantum-classical workflow for drug discovery that came within 4% of a classical benchmark, reaching the chemical-accuracy threshold for modeling a complex enzyme active site.

 

The test combined QC Ware’s Promethium quantum chemistry platform with IonQ’s Forte trapped-ion quantum computer through Amazon Braket. The companies modeled the heme active site of cytochrome P450nor, a nitric oxide reductase in the cytochrome P450 superfamily, which includes enzymes responsible for most human drug metabolism.

 

The workflow calculated electrostatic interaction energy to within 0.5 kcal/mol — roughly 4% — of classical benchmark results. That is well inside the commonly cited 1 kcal/mol threshold for chemical accuracy and, according to QC Ware, delivered more than twice the accuracy of the standard classical mean-field approach.

 

The result could have implications for pharmaceutical research, where accurately calculating how strongly drug candidates interact with complex metal-containing enzyme sites can help researchers rank compounds and identify potential metabolic risks earlier.

 

"Running the same hybrid workflow on IonQ's trapped-ion architecture, following our recent demonstration on other quantum hardware, shows that Promethium's approach to combining classical and quantum computing is not tied to a single type of quantum hardware," said Kin-Joe Sham, Co-Founder and COO at QC Ware. "We believe this hardware-agnostic approach gives researchers flexibility as quantum computing continues to mature."

 

The demonstration began with Promethium preprocessing a 115-atom model containing more than 1,000 molecular orbitals. The platform reduced the strongly correlated portion of the system to a four-orbital active space, which was mapped onto eight qubits for measurement on IonQ Forte.

 

IonQ's all-to-all qubit connectivity allowed the required two-qubit entangling gates to run without the routing overhead associated with architectures with limited connectivity. The resulting measurements were then returned to Promethium, where the final interaction energies were calculated classically.

 

"Every month spent advancing a drug candidate on flawed metabolic data is wasted time and mounting risk," said Scott Millard, Chief Business Officer at IonQ. "QC Ware and IonQ have shown that hybrid quantum-classical workflows can predict certain binding behavior accurately enough for discovery teams to confidently rank candidates and catch toxicity risks early. We believe that's real quantum impact on real health outcomes. I can't wait to see what this partnership delivers."

 

The companies said the demonstration, supported in part by Amazon Web Services cloud-compute credits, highlights a cloud-based model in which GPU-accelerated classical computing can be combined with quantum resources through Amazon Braket.

 

QC Ware said Promethium is designed to extend quantum chemistry calculations to larger molecular systems and compound libraries. For selected workloads, the company says its GPU-native platform can perform calculations up to 20 times faster than conventional CPU-based density functional theory platforms.