Título del libro: Drug Design Using Machine Learning
Título del capítulo: Molecular recognition and machine learning to predict protein-ligand interactions
ANGEL RAMON HERNANDEZ MARTINEZ;
Autores externos:
Idioma:
Año de publicación:
2022
Palabras clave:
Mathematical models; Molecular docking; Scoring function; Search algorithm
Resumen:
Molecular recognition is part of several chemical-biological processes, and is the interaction between macromolecules (such as proteins and ligands) through noncovalent bonds. This phenomenon has been extensively studied for developing new drugs. Molecular modeling is an affordable method (compared with laboratory experiments) for predicting which macromolecules may interact and, through molecular docking, which will form a stable complex. Molecular docking has two main components: (1) search algorithm and (2) scoring function. The search algorithm studies the conformational space of the ligand at the binding site. The scoring function is a mathematical model that evaluates the interaction energy of each complex, and it could be empirical by using databases of ligand-protein complexes. Results of the search algorithm are satisfactory compared with experimental data, but the scoring function still must improve its performance. Due to the complexity of analysis and management of databases, accurate predictions are difficult to obtain. Machine learning can contribute to achieve better results for predicting macromolecular interactions. Computational predictions of the interaction between macromolecules complexes enhance the development of applied technology in medicine. © 2022 Scrivener Publishing LLC.
Entidades citadas de la UNAM:
ISBN:
9781394167258
Editorial:
wiley Nombre de la publicación:
Molecular recognition and machine learning to predict protein-ligand interactions
Página inicial:
1
Página final:
20
