Dataset Information

Exploring the potential of protein-based pharmacophore models in ligand pose prediction and ranking.

ABSTRACT: Protein-based pharmacophore models derived from protein binding site atoms without the inclusion of any ligand information have become more popular in virtual screening studies. However, the accuracy of protein-based pharmacophore models for reproducing the critical protein-ligand interactions has never been explicitly assessed. In this study, we used known protein-ligand contacts from a large set of experimentally determined protein-ligand complexes to assess the quality of the protein-based pharmacophores in reproducing these critical contacts. We demonstrate how these contacts can be used to optimize the pharmacophore generation procedure to produce pharmacophore models that optimally cover the known protein-ligand interactions. Finally, we explored the potential of the optimized protein-based pharmacophore models for pose prediction and pose rankings. Our results demonstrate that there are significant variations in the success of protein-based pharmacophore models to reproduce native contacts and consequently native ligand poses dependent on the details of the pharmacophore generation process. We show that the generation of optimized protein-based pharmacophore models is a promising approach for ligand pose prediction and pose rankings.

SUBMITTER: Hu B

PROVIDER: S-EPMC3725268 | biostudies-literature | 2013 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Exploring the potential of protein-based pharmacophore models in ligand pose prediction and ranking.

Hu Bingjie B Lill Markus A MA

Journal of chemical information and modeling 20130513 5

Protein-based pharmacophore models derived from protein binding site atoms without the inclusion of any ligand information have become more popular in virtual screening studies. However, the accuracy of protein-based pharmacophore models for reproducing the critical protein-ligand interactions has never been explicitly assessed. In this study, we used known protein-ligand contacts from a large set of experimentally determined protein-ligand complexes to assess the quality of the protein-based ph ...[more]

PMID: 23621564

Similar Datasets

Project description:Protein-ligand interactions are increasingly profiled at high-throughput, playing a vital role in lead compound discovery and drug optimization. Accurate prediction of binding pose and binding affinity constitutes a pivotal challenge in advancing our computational understanding of protein-ligand interactions. However, inherent limitations still exist, including high computational cost for conformational search sampling in traditional molecular docking tools, and the unsatisfactory molecular representation learning and intermolecular interaction modeling in deep learning-based methods. Here we propose a geometry-aware attention-based deep learning model, GAABind, which effectively predicts the pocket-ligand binding pose and binding affinity within a multi-task learning framework. Specifically, GAABind comprehensively captures the geometric and topological properties of both binding pockets and ligands, and employs expressive molecular representation learning to model intramolecular interactions. Moreover, GAABind proficiently learns the intermolecular many-body interactions and simulates the dynamic conformational adaptations of the ligand during its interaction with the protein through meticulously designed networks. We trained GAABind on the PDBbindv2020 and evaluated it on the CASF2016 dataset; the results indicate that GAABind achieves state-of-the-art performance in binding pose prediction and shows comparable binding affinity prediction performance. Notably, GAABind achieves a success rate of 82.8% in binding pose prediction, and the Pearson correlation between predicted and experimental binding affinities reaches up to 0.803. Additionally, we assessed GAABind's performance on the severe acute respiratory syndrome coronavirus 2 main protease cross-docking dataset. In this evaluation, GAABind demonstrates a notable success rate of 76.5% in binding pose prediction and achieves the highest Pearson correlation coefficient in binding affinity prediction compared with all baseline methods.

Dataset Information

Exploring the potential of protein-based pharmacophore models in ligand pose prediction and ranking.

Publications

Exploring the potential of protein-based pharmacophore models in ligand pose prediction and ranking.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets