Framework for investigating the current–voltage topographies prediction model of quantum-confined metallic gate enclosed biomolecular nanoribbon based on DFT and machine learning

Abstract This study presents the optimal prediction model for a quantum-confined adenine molecular nanoribbon, developed using a hybrid approach that integrates density functional theory with machine learning. The proposed framework enables precise prediction of electrical characteristics, including a visual representation of the current–voltage ( I – V ) response. Various electronic properties of the quantum-confined biomolecular nanoribbon are predicted through a regression learner model. To generate and supply training data for constructing future prediction models, the approach incorporates a learnable real-space Hamiltonian method alongside k -point sampling. The best-performing regression learner achieves an accuracy of approximately 89% for this quantum-confined adenine nanoribbon structure. Overall, this method delivers faster and more accurate electronic characterization of quantum-confined, bio-inspired nanoscale systems.