Molecular docking with Gaussian Boson Sampling.

Area of Science:

• Quantum Computing
• Computational Chemistry
• Drug Discovery

Background:

• Gaussian Boson Samplers (GBS) are photonic quantum devices with potential for complex computations.
• A key challenge is identifying practical applications for near-term quantum technologies.
• Molecular docking is crucial for pharmaceutical drug design, predicting how molecules bind to targets.

Purpose of the Study:

• To demonstrate the utility of Gaussian Boson Samplers for predicting molecular docking configurations.
• To develop a quantum approach for identifying stable molecular docking poses.
• To explore the integration of quantum computation with classical algorithms for drug design.

Main Methods:

• Reducing the molecular docking problem to finding the maximum weighted clique in a graph.
• Programming Gaussian Boson Samplers to sample large-weight cliques, representing stable docking configurations.
• Simulating the GBS approach, accounting for photon losses.
• Benchmarking the method using a ligand binding to tumor necrosis factor-α converting enzyme.

Main Results:

• Gaussian Boson Samplers can successfully predict molecular docking configurations with high probability.
• The quantum approach demonstrates robustness even in the presence of photon losses.
• Outputs from the GBS device can improve the performance of classical docking algorithms.
• The method was benchmarked on a biologically relevant protein-ligand interaction.

Conclusions:

• Gaussian Boson Samplers offer a promising quantum advantage for molecular docking in drug discovery.
• This work bridges quantum computing capabilities with a critical pharmaceutical research problem.
• The developed quantum-classical hybrid approach has the potential to accelerate drug design pipelines.