Staats- und Universitätsbibliothek Hamburg Carl von Ossietzky
Erscheinungsjahr:
2023
Medientyp:
Text
Schlagworte:
structure-based virtual screening
ligand-based virtual screening
machine learning
benchmarking
virtual screening strategies
004: Informatik
ddc:004:
Beschreibung:
The pharmaceutical industry has been making substantial use of computer-aided drug discovery methods, particularly virtual screening approaches, for the identification and optimization of lead compounds. In recent years, numerous structure-based and ligand-based virtual screening approaches have been developed and evaluated to determine their performance and technical setup. Each approach, however, has its benefits and limitations, and no virtual screening method consistently outperforms the others. In addition, experimentally-validated data regarding active ligands and protein structures is scarce and further explorations are required to take full advantage of them. The evaluation and combination of different methods is a well-established strategy for increasing the success rates of virtual screening campaigns. In this thesis, we sought to employ and evaluate different virtual screening protocols as well as characterize and develop various combinations of strategies based on structure-based and ligand-based methods.