QuantumMed: Leveraging Quantum Genetic Programming for Advanced Drug Discovery

Abstract

Drug development in bioinformatics and bioengineering is a complicated and ambitious procedure aiming at discovering new pharmacological candidates. Despite technological breakthroughs and improved understanding of biological processes, drug development remains an expensive, complex, and inefficient process with a consistently low success rate. The pharmaceutical industry sometimes views the discovery of new pharmaceuticals as a business with substantial commercial or public health ramifications. To handle the growing complexity of chemical databases, new knowledge discovery methods are necessary. The computing cost of these methods rises directly proportional to the number of compounds. Investigating the relationship between chemical compounds and their biological or chemical properties is a crucial step in this process. Conventional methods, such as classical genetic programming and neural networks, frequently generate erroneous predictions. These approaches frequently necessitate substantial parameter fine-tuning and sophisticated transformations of predictor or result variables, reducing their effectiveness. In this study, we used a Quantum Genetic Programming (QGP) model to significantly improve prediction accuracy. QGP develops linear equations, providing a more precise method for calculating the degree of chemical properties. The results reveal that QGP greatly outperforms conventional genetic programming, indicating a prospective advance in drug development.