Publications by Adrian Schreyer

Syndicate content NCBI pubmed
NCBI: db=pubmed; Term=schreyer, adrian
Updated: 4 days 7 hours ago

Does a more precise chemical description of protein-ligand complexes lead to more accurate prediction of binding affinity?

Mon, 14/08/2017 - 12:39
Related Articles

Does a more precise chemical description of protein-ligand complexes lead to more accurate prediction of binding affinity?

J Chem Inf Model. 2014 Mar 24;54(3):944-55

Authors: Ballester PJ, Schreyer A, Blundell TL

Abstract
Predicting the binding affinities of large sets of diverse molecules against a range of macromolecular targets is an extremely challenging task. The scoring functions that attempt such computational prediction are essential for exploiting and analyzing the outputs of docking, which is in turn an important tool in problems such as structure-based drug design. Classical scoring functions assume a predetermined theory-inspired functional form for the relationship between the variables that describe an experimentally determined or modeled structure of a protein-ligand complex and its binding affinity. The inherent problem of this approach is in the difficulty of explicitly modeling the various contributions of intermolecular interactions to binding affinity. New scoring functions based on machine-learning regression models, which are able to exploit effectively much larger amounts of experimental data and circumvent the need for a predetermined functional form, have already been shown to outperform a broad range of state-of-the-art scoring functions in a widely used benchmark. Here, we investigate the impact of the chemical description of the complex on the predictive power of the resulting scoring function using a systematic battery of numerical experiments. The latter resulted in the most accurate scoring function to date on the benchmark. Strikingly, we also found that a more precise chemical description of the protein-ligand complex does not generally lead to a more accurate prediction of binding affinity. We discuss four factors that may contribute to this result: modeling assumptions, codependence of representation and regression, data restricted to the bound state, and conformational heterogeneity in data.

PMID: 24528282 [PubMed - indexed for MEDLINE]