Titel
Applications of the Novel Quantitative Pharmacophore Activity Relationship Method QPhAR in Virtual Screening and Lead-Optimisation
... show all
Abstract
Pharmacophores are an established concept for the modelling of ligand–receptor interactions based on the abstract representations of stereoelectronic molecular features. They became widely popular as filters for the fast virtual screening of large compound libraries. A lot of effort has been put into the development of sophisticated algorithms and strategies to increase the computational efficiency of the screening process. However, hardly any focus has been put on the development of automated procedures that optimise pharmacophores towards higher discriminatory power, which still has to be done manually by a human expert. In the age of machine learning, the researcher has become the decision-maker at the top level, outsourcing analysis tasks and recurrent work to advanced algorithms and automation workflows. Here, we propose an algorithm for the automated selection of features driving pharmacophore model quality using SAR information extracted from validated QPhAR models. By integrating the developed method into an end-to-end workflow, we present a fully automated method that is able to derive best-quality pharmacophores from a given input dataset. Finally, we show how the QPhAR-generated models can be used to guide the researcher with insights regarding (un-)favourable interactions for compounds of interest.
Stichwort
pharmacophorepharmacophore modellingquantitative pharmacophoreQSARmachine learningpharmacophore optimisationNeuroDeRisk
Objekt-Typ
Sprache
Englisch [eng]
Persistent identifier
Erschienen in
Titel
Pharmaceuticals
Band
15
Ausgabe
9
ISSN
1424-8247
Erscheinungsdatum
2022
Publication
MDPI AG
Projekt
Kod / Identifikator
821528
Erscheinungsdatum
2022
Zugänglichkeit
Rechteangabe
© 2022 by the authors

Herunterladen

Universität Wien | Universitätsring 1 | 1010 Wien | T +43-1-4277-0