Quantum-machine-assisted Drug Discovery

TL;DR

This paper examines integrating quantum computing across the drug development cycle to accelerate and enhance workflows and rigorous decision-making and highlights quantum approaches for molecular simulation, drug-target interaction prediction, and optimizing clinical trials.

Abstract

Drug discovery is lengthy and expensive, with traditional computer-aided design facing limits. This paper examines integrating quantum computing across the drug development cycle to accelerate and enhance workflows and rigorous decision-making. It highlights quantum approaches for molecular simulation, drug-target interaction prediction, and optimizing clinical trials. Leveraging quantum capabilities could accelerate timelines and costs for bringing therapies to market, improving efficiency and ultimately benefiting public health.

Figures (4)

• Figure 1: Pipeline of drug discovery and development. Drug discovery prioritizes a small set of novel structures with desirable properties, followed by pre-clinical studies in vivo and phased clinical development to evaluate safety and efficacy. The stage durations and attrition rates illustrated here reflect representative ranges from regulatory and research bodies such as the U.S. Food and Drug Administration (FDA) and the National Institutes of Health (NIH) fda_devpathfda_pdufa_goalsnih_ncats_timelinectgov_stats. The magnitude of small-molecule chemical space and the early-stage funnel from very large libraries to tractable candidate sets are supported by peer-reviewed surveys and modern ultra-large screening exemples polishchuk2013estimationtingle2023zinclyu2019ultragorgulla2020openzhou2024artificial.
• Figure 2: A common way to visualize the state of a single-qubit is to parametrize it as $|\Psi\rangle = cos(\theta/2)|0\rangle + e^{i\phi}sin(\theta/2)|1\rangle$ where the angles $\theta$ , $\phi$ map the state onto a point on the surface of a sphere, known as the Bloch sphere. The north and south poles, $|0\rangle$ and $|1\rangle$, represent the “classical states” (or computational basis states) and denote the bits 0, 1 used in a classical computer.
• Figure 3: Overview of quantum computing integration in drug development and clinical trial optimization. (a): Classical pre-processing pipeline integrating diverse data sources for eligibility criteria determination, including experimental structure analysis via microscopy, patient medical records, laboratory test results, and genomic data. (b): Comparative analysis of classical versus quantum approaches in drug discovery, contrasting traditional high-throughput screening methods that search through approximately $10^9$ known molecules with quantum-enabled de novo drug design capable of exploring a vastly larger chemical space of $10^{60}$ molecules. The quantum approach uses quantum algorithms to navigate this expanded molecular space more efficiently than classical methods. (c): Detailed quantum algorithm implementation showing the progression from molecular structure to quantum circuit design, incorporating VQE for calculating molecular properties like excitation energies and excited states. This demonstrates how quantum computing can provide more accurate modeling of molecular behavior compared to classical approximation. (d): Quantum-classical hybrid optimization framework for clinical trial management, where quantum circuits generate trial site configurations while working in conjunction with classical optimizers to calculate loss functions. The framework optimizes across multiple parameters, including enrollment rates and costs, visualized in a matrix of trial sites with varying performance metrics.
• Figure 4: Schematic illustration of QFL. Multiple clients hold private data that can not be leaked to or shared with the outside, while the QFL scheme enables the training of a public model without uploading their private data.