Over several decades, a variety of computational methods for drug discovery have been proposed and applied in practice. With the accumulation of data and the development of machine learning methods... Show moreOver several decades, a variety of computational methods for drug discovery have been proposed and applied in practice. With the accumulation of data and the development of machine learning methods, computational drug design methods have gradually shifted to a new paradigm, i.e. deep learning methods have attracted particular interest in drug design. In this study, a new deep learning-based method (DrugEx) was proposed to design de novo drug-like molecules. It was proven that candidate molecules designed by DrugEx had a larger chemical diversity, and better covered the chemical space of known ligands. In order to address the issue of polypharmacology, the DrugEx algorithm was updated with multi-objective optimization towards multiple targets. The results of its application demonstrated the generation of compounds with a diverse predicted selectivity profile toward multiple targets, offering the potential of high efficacy and lower toxicity. In order to improve its generality, DrugEx was further updated to have the capability of designing molecules based on given scaffolds. We extended the architecture of Transformer to deal with each molecule as a graph. As a proof, its effectiveness in that 100% valid molecules are generated and most of them had predicted high affinity towards A2AAR with given scaffolds. Moreover, GenUI was developed as a visualizion software platform that makes it possible to integrate molecular generators within a feature-rich graphical user interface to facilitate collaboration in the disparate communities interested in computer-aided drug discovery.These studies highlight the overwhelming power of AI methods in drug discovery. Show less
One of the main problems of drug design is that it is quite hard to discover compounds that have all the required properties to become a drug (efficacy against the disease, good biological... Show moreOne of the main problems of drug design is that it is quite hard to discover compounds that have all the required properties to become a drug (efficacy against the disease, good biological availability, low toxicity). This thesis describes the use of data mining and interactive evolutionary algorithms to design novel classes of molecules. Using data mining, we split a 250,000 compound database into ring systems, substituents and linkers. We then counted the occurrence of the different fragments, as well as their co-occurrence. Our resulting lists of common and uncommon chemical substructures and substructure combinations can be used to increase the diversity of drug screening libraries and hence increase their chance to yield new drugs. We also developed a computer program, the Molecule Evoluator. This program uses an interactive evolutionary algorithm to propose novel molecules or molecule modifications. Using the Molecule Evoluator, our chemists were able to discover three novel classes of compounds, resulting in the synthesis of eight new compounds. Four of these proved to bind to biogenic amine targets such as the norepinephrine transport protein and the alpha-adrenergic receptors. So, our computer methods offer inspiration to chemists, helping them to get new ideas for drug molecules. Show less
