Quantum-Enhanced Drug Discovery Using HybridAI and Quantum Computing for Molecular Property Prediction and Virtual Screening

Abstract

: The trajectory of pharmaceutical development is increasingly unsustainable due to escalating costs and protracted timelines. Traditional computational methods, fundamentally constrained by exponential scaling when simulating quantum mechanical effects, face accuracy limitations when dealing with complex molecular systems. This paper introduces the Quantum Enhanced Drug Discovery framework, a novel paradigm that rigorously integrates advanced classical machine learning with emerging Quantum Computing capabilities. The core of our approach is a Hybrid Artificial Intelligence model that marries a Graph Convolutional Network with quantum feature generation via the Variational Quantum Eigensolver. This fusion allows for the construction of a powerful molecular representation that captures both topological structure and accurate electronic behavior. We demonstrate the application of this framework to Quantitative Structure Activity Relationship modeling for the precise prediction of Absorption, Distribution, Metabolism, Excretion, and Toxicity properties and lead compound prioritization during virtual screening. Our results indicate that the quantum feature vector significantly improves predictive performance, offering a viable route to overcome computational bottlenecks and revolutionize the efficiency of early stage drug discovery.