• mutations
• Molecular Dynamics simulations
• AlphaFold2
• Protein-ligand binding affinity

Proteins are important macro-molecules of living creatures as they carry out important functions of organisms. To fulfil their tasks, proteins bind with other proteins and smaller molecules called ligands. This binding can be characterized by the \textit{binding affinity}, which can be considered as the strength of the interaction between proteins and their partners. When mutations occur within a protein, one or more amino acids of the original (wild type) protein are replaced by different amino acids. This replacement can affect the binding affinity of the protein with its ligands. To address this problem, this thesis proposes an automatized computational workflow to compute the difference in the binding free energy between the wild type the mutated proteins. The workflow consists of 3 main steps. The first step is to predict the structure of the mutated protein with a state-of-the-art protein structure prediction tool - AlphaFold2. The second step is to run a Molecular Dynamics simulation to get the behavior of proteins in contact with ligands. The binding free energy of proteins and ligands is then calculated in the final step. By comparing the binding free energy of the wild type protein and the mutated protein, the effect of the mutation on the binding affinity can be inferred. The workflow is tested with 48 mutations of 29 proteins. A clear result can be drawn for 28 mutations, in which, 6 are incorrect and 22 are correct. This makes up the accuracy of 78.57%.