Deep neural network affinity model for BACE inhibitors in D3R Grand Challenge 4

Abstract

Drug Design Data Resource (D3R) Grand Challenge 4 (GC4) offered a unique opportunity for designing and testing novel methodology for accurate docking and affinity prediction of ligands in an open and blinded manner. We participated in the beta-secretase 1 (BACE) Subchallenge which is comprised of cross-docking and redocking of 20 macrocyclic ligands to BACE and predicting binding affinity for 154 macrocyclic ligands. For this challenge, we developed machine learning models trained specifically on BACE. We developed a deep neural network (DNN) model that used a combination of both structure and ligand-based features that outperformed simpler machine learning models. According to the results released by D3R, we achieved a Spearman's rank correlation coefficient of 0.43 (7) for predicting the affinity of 154 ligands. We describe the formulation of our machine learning strategy in detail. We compared the performance of DNN with linear regression, random forest, and support vector machines using ligand-based, structurebased, and combining both ligand and structure-based features. We compared different structures for our DNN and found that performance was highly dependent on fine optimization of the regularization hyperparameter, alpha. We also developed a novel metric of ligand threedimensional similarity inspired by crystallographic difference density maps to match ligands without crystal structures to similar ligands with known crystal structures. This report demonstrates the detailed parameterization and careful data training and implementation necessary to obtain strong performance with more complex machine learning methods. Our DNN approach tied for fourth in predicting BACE-ligand binding affinities.

Pepsin-like aspartic proteases (PAPs) as model systems for combining biomolecular simulation with biophysical experiments

“…If Tyr residue shows Fermi resonance in Raman spectra that gives an indication of the rotational degrees of freedom associated with the Tyr side-chain. 103 docking, [111][112][113] free energy calculations 114 (e.g. FEP, ABFE, MM/ PBSA or MM/GBSA) and biochemical assay can then be used to identify BACE-1 inhibitors (in broad sense this concept is applicable for other PAP inhibitors e.g.…”

On the Conformational Dynamics of β-Amyloid Forming Peptides: A Computational Perspective

“…screened PubChem compounds library and found wgx-50 as a potential inhibitor of Aβ-42, the synergistic effects of with gold nanoparticles induced significant inhibition of Aβ 1-42 relative to that induced by wgx-50 alone. Wang and Ng (2019) developed five different machine learning models ranging in complexity from linear regression to a deep neural network.…”

“…The deep neural network trained specifically on BACE ligands performed best for affinity prediction (Wang and Ng, 2019). Current applications of deep learning in drug development have the potential to facilitate the development of a drug for AD because of several reasons.…”

RORγ Structural Plasticity and Druggability

“…Therefore, the plasticity of the pocket needs to be considered in VS to avoid missing some important hits. The state-of-the-art docking methods incorporate multiple model templates and machine learning scoring functions to markedly improve the effectiveness of predicting ligand activity and even the quantitative predictions of binding affinity [ 52 ].…”