Improvement of protein-ligand binding affinity prediction using machine learning techniques
Autor: | Hernandez, Gabriela, Iglesias, Jelisa, Saen-oon, Suwipa |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2015 |
Předmět: |
Medicaments -- Disseny
Drug discovery Protein-ligand binding Ligands (Biochemistry) Lligands (Bioquímica) Ciències de la salut::Medicina::Farmacologia [Àrees temàtiques de la UPC] Proteïnes -- Fixació Scoring functions Machine learning Protein binding High performance computing Drugs -- Design Càlcul intensiu (Informàtica) Informàtica::Arquitectura de computadors [Àrees temàtiques de la UPC] |
Zdroj: | UPCommons. Portal del coneixement obert de la UPC Universitat Politècnica de Catalunya (UPC) |
Popis: | Predicting protein-ligand binding affinities constitutes a key computational method in the early stages of the drug discovery process. Molecular docking programs attempt to predict them by using mathematical approximations, namely, scoring functions. In the last years, several scoring functions have been developed, encompassing different terms, from electrostatic forces to protein-ligand interaction fingerprints and beyond. However, it has been noticed that usually each individual scoring function cannot be generalized and its predictive power is arguable. The aim of this study is to improve the binding affinity prediction by finding potential models to combine ten different scoring functions, exploiting machine learning techniques. |
Databáze: | OpenAIRE |
Externí odkaz: |