IJAIEM

Rama Krishna Mani Kanta Yalla, Thirusubramanian Ganesan, Mohanarangan Veerapperumal Devarajan, Akhil Raj Gaius Yallamelli, Vijaykumar Mamidala, Aiswarya RS

Abstract

Development of efficient predictive models became essential for computational drug discovery due to its fast progress in identifying drug candidates. The proposed research developed a hybrid framework combining quantum computing with machine learning and cloud computing to refine drug-target interaction detection capabilities. The research combines the IBM Quantum platform through Qiskit to execute Variational Quantum Eigen solver algorithm which generates quantum-derived molecular descriptors that represent quantum mechanical features in drug-like molecules. The quantum-derived descriptors join classical molecular descriptors as input features for the subsequent Extreme Gradient Boosting model. XGBoost effectively handles these high- dimensional features by using them to identify drug-target binding affinities and determine drug candidate priorities. The workflow operates from IBM Cloud as a deployment system to provide effortless quantum simulation capabilities and instantan