Year: 2022
Type: dissertation
Abstract: Computer-aided drug discovery (CADD) methods are now routinely used in the preclinical phase of drug development. Powerful high-performance computing facilities and the extremely fast CADD methods constantly scale up the coverage of drug-like chemical space achievable in rational drug development. In this thesis, CADD approaches were applied to address several early-phase drug discovery problems. Namely, small molecule binding site detection on a novel target protein, virtual screening (VS) of molecular databases, and characterization of small molecule interactions with metabolic enzymes were studied. Various CADD methods, including molecular dynamics (MD) simulations in mixed solvents, molecular docking, and binding free energy calculations, were employed. Co-solvent MD simulations detected biologically relevant binding sites and provided guidance for screening potential protein-protein interaction inhibitors for a crucial protein of the SARS-CoV-2. VS with fragment- and negative image-based (F-NIB) models identified three active and structurally novel inhibitors of the putative drug target phosphodiesterase 10A. MD simulations and docking provided detailed insights on the effects of active site structural flexibility and variation on the binding and resultant metabolism of small molecules with the cytochrome P450 enzymes. The results presented in this thesis contribute to the increasing evidence that supports employment and further development of CADD approaches in drug discovery. Ultimately, rational drug development coupled with CADD may enable higher quality drug candidates to the human studies in the future, reducing the risk of financially and temporally costly clinical failure. KEYWORDS: Structure-based drug development, Computer-aided drug discovery (CADD), Molecular dynamics (MD) simulation, Virtual screening (VS), Fragmentand negative image-based (F-NIB) model, Structure-activity relationship (QSAR), Cytochrome P450 ligand binding prediction
Source: UTUPub (University of Turku)
Author: Jokinen, Elmeri
Language: English
Subfield: Drug Discovery
Domain: Life Sciences
Keywords: computational drug discovery, molecular dynamics simulation, virtual screening, fragment-based virtual screening, binding site identification, protein-protein interaction inhibitors, SARS-CoV-2, PDE10A, cytochrome P450, drug metabolism, structure-activity relationship, binding free energy, mixed solvents, molecular docking
Open Access status: green
OpenAlex ID: W7058476961
