Title
Quantum Machine Learning in Drug Discovery
Abstract
Drugs act through interactions with target proteins, regulating biological functions to prevent and treat disease. However, developing a new drug is costly, often takes more than a decade, and has very high failure rates. Machine learning and deep learning have contributed significantly to improving this process. They remain complex, though, and require large volumes of high-quality data and substantial computational power. They also struggle to describe interactions accurately at the molecular and quantum levels.
Quantum computing exploits phenomena such as superposition and entanglement. This opens new possibilities for processing complex information and simulating molecular systems. Combining quantum computing with machine learning has given rise to the rapidly growing field of quantum machine learning. This field is proposed as a promising way to reduce the cost and time of drug development.
The fundamental principles of quantum computing and quantum machine learning are presented, along with the limitations of today’s noisy intermediate-scale quantum (NISQ) devices. Hybrid quantum-classical approaches, which combine classical and quantum processing to overcome hardware limitations, are examined. So are their applications across the various stages of drug discovery. Particular attention is given to drug–protein binding affinity prediction, as a representative example of the research carried out at the Department of Digital Systems, University of Thessaly. Key issues are also discussed, including data representation, quantum encoding, and the reliable benchmarking of classical and quantum models.
Finally, the prospects of the field are analyzed, together with the pharmaceutical industry’s growing interest and future research directions. A central question remains whether quantum advantage exists, and how it can be leveraged in real-world problems in the life sciences.
Short CV
Maria Avramouli is a member of the Laboratory Teaching Staff at the Department of Digital Systems, University of Thessaly. She holds a Diploma in Computer Engineering and Informatics from the Polytechnic School of the University of Patras, an MSc in Information and Communication Systems from the Open University of Cyprus, and a PhD in Quantum Machine Learning (QML). She has extensive professional, teaching, and research experience in software development, machine learning, and quantum computing, having previously worked as a software engineer and as a secondary education teacher. Her technical expertise includes Python, C/C++, Java, PHP, and JavaScript, as well as quantum computing and machine learning frameworks such as PennyLane, Qiskit, and PyTorch. Her research interests focus on quantum and classical machine learning, the development of hybrid quantum–classical models and algorithms, and their application to complex computational problems, with particular emphasis on drug discovery and computational biomedical research. She has participated in EU-funded research projects, authored and co-authored scientific publications in peer-reviewed journals and international conferences, and serves as a reviewer and member of scientific and technical committees for international conferences.