Integrating machine learning (ML) into drug discovery has ushered in a new era of innovation, dramatically enhancing the efficiency and precision of identifying and developing new therapeutics. This review provides a comprehensive analysis of the current applications of machine learning in drug discovery, focusing on its transformative impact across various stages of the drug development pipeline. We delve into key ML methodologies, including supervised and unsupervised learning, neural networks, and reinforcement learning, examining their underlying principles and specific contributions to drug discovery processes. By exploring case studies and recent advancements, this review illustrates how ML algorithms have been utilized to predict drug-target interactions, optimize drug design, and streamline clinical trial processes. Furthermore, we discuss the challenges and limitations of implementing ML techniques in this field and highlight emerging trends and future directions. This review aims to offer researchers a thorough understanding of ML's potential to revolutionize drug discovery and equip them with the insights needed to leverage these technologies effectively.
Integrating machine learning (ML) into drug discovery has ushered in a new era of innovation, dramatically enhancing the efficiency and precision of identifying and developing new therapeutics. This review provides a comprehensive analysis of the current applications of machine learning in drug discovery, focusing on its transformative impact across various stages of the drug development pipeline. We delve into key ML methodologies, including supervised and unsupervised learning, neural networks, and reinforcement learning, examining their underlying principles and specific contributions to drug discovery processes. By exploring case studies and recent advancements, this review illustrates how ML algorithms have been utilized to predict drug-target interactions, optimize drug design, and streamline clinical trial processes. Furthermore, we discuss the challenges and limitations of implementing ML techniques in this field and highlight emerging trends and future directions. This review aims to offer researchers a thorough understanding of ML's potential to revolutionize drug discovery and equip them with the insights needed to leverage these technologies effectively.
Cancer is highly complex, including cross-talks between signalling pathways and multiple genes. Fortunately, successful anti-cancer drugs, such as cisplatin, are promising to treat cancer. Despite its clinical success, cisplatin presents several issues, including neurotoxicity, ototoxicity, nephrotoxicity and drug resistance. These limitations of the successful anti-cancer cisplatin can be overcome with the help of novel supramolecular drug delivery systems. This research aims to optimise novel supramolecular drug delivery systems – metallacage (M2L4}(M = metal, L = ligand) – that can be used to encapsulate cisplatin to protect it from metabolism. Specifically, we investigated the shape and stability of ten metallacages using different metals: Pt2+, Pd2+, Ni2+, Cu2+ and Au3+ using Density Functional Theory (DFT) methods to encapsulate one or two cisplatin molecules. Thus, we applied WebMo running PBE0 and Hartree Fock as theories, with LanL2DZ as a basis set. The results demonstrate that PBE0/LanL2DZ is the best method to describe the cages. The results show that Endo-N Ni2L4 and Endo-N Cu2L4 cages are the best option to encapsulate one and two cisplatin, while the least suitable system is Endo-C Pt2L4 and Endo-C Au2L4 for cisplatin encapsulation. Furthermore, endo-C cages are farther from ground state energy because of their higher energy than endo-N cages. Therefore, endo-N cages are superior to endo-C cages for the encapsulation of cisplatin.