Využití umělé inteligence při návrhu proteinových struktur a její dopad na efektivitu vývoje léčiv

This bachelor’s thesis deals with the use of artificial intelligence in protein structure prediction and its impact on the efficiency of drug development. The aim of the thesis is to analyze the benefits of deep learning methods in predicting the three-dimensional structure of proteins, compare them with traditional experimental and computational approaches, and evaluate their significance for contemporary biological and pharmaceutical research. The thesis first introduces the biological foundations of proteins, their structure, and their importance for the functioning of living organisms. It then describes traditional methods for determining protein structures, such as X-ray crystallography, cryo-electron microscopy, and nuclear magnetic resonance, and also summarizes the classical computational approaches used before the emergence of modern AI models. The next part of the thesis focuses on the principles of machine learning and deep learning and on their application in bioinformatics. Special attention is paid to the development of modern protein structure prediction methods over the last decade and to specific models, especially AlphaFold, AlphaFold 3, RoseTTAFold, and ESMFold. For these models, their principles, benefits, limitations, and practical applications are analyzed. The analytical part of the thesis evaluates the impact of these methods on the drug development process, especially in terms of research speed, therapeutic target selection, virtual screening, and the overall efficiency of the early stages of development. The thesis also discusses the limitations of current approaches, the need for experimental validation, and the main directions of future development in the field of AI-based protein structure prediction and drug design.