Quantum computing applications in drug discovery

Area of Science:

• Computational chemistry
• Quantum computing applications
• Drug discovery

Background:

• Virtual screening (deep learning, molecular docking) and molecular dynamics are key in drug discovery but face trade-offs.
• Noisy Intermediate-Scale Quantum (NISQ) era presents unique challenges and opportunities for quantum integration.

Purpose of the Study:

• To explore the integration of quantum computing into early drug discovery workflows.
• To identify realistic roles for quantum computing as modular coprocessors within classical frameworks.
• To define criteria for evaluating the progress of quantum-enhanced computational strategies.

Main Methods:

• Modular quantum coprocessing for deep learning screening (representation learning, feature extraction, generative models).
• Quantum integration in docking screening (site recognition, pose search, flexible docking).
• Quantum applications in molecular dynamics (ab initio, trajectory propagation, excited state dynamics).

Main Results:

• Quantum modules can enhance specific subroutines in deep learning, docking, and molecular dynamics.
• Most large-scale sampling in molecular dynamics still relies on classical methods.
• The effectiveness of quantum modules depends on repeatable decision-making gains within resource constraints.

Conclusions:

• Quantum computing's near-term role is modular coprocessing, not complete replacement of classical methods.
• Evaluation criteria include classical baselines, ranking reliability, resource reporting, and downstream decision impact.
• Successful quantum integration requires demonstrating tangible benefits under NISQ limitations.