AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery

By Izhar Wallach, Michael Dzamba, Abraham Heifets. 2015

596 citations recorded by the source corpus.

Deep convolutional neural networks comprise a subclass of deep neural networks (DNN) with a constrained architecture that leverages the spatial and temporal structure of the domain they model. Convolutional networks achieve the best predictive performance in areas such as speech and image recognition by hierarchically composing simple local features into complex models. Although DNNs have been used in drug discovery for QSAR and ligand-based bioactivity predictions, none of these models have benefited from this powerful convolutional architecture. This paper introduces AtomNet, the first structure-based, deep convolutional neural network designed to predict the bioactivity of small molecules for drug discovery applications. We demonstrate how to apply the convolutional concepts of feature locality and hierarchical composition to the modeling of bioactivity and chemical interactions. In further contrast to existing DNN techniques, we show that AtomNet's application of local convolutional filters to structural target information successfully predicts new active molecules for targets with no previously known modulators. Finally, we show that AtomNet outperforms previous docking approaches on a diverse set of benchmarks by a large margin, achieving an AUC greater than 0.9 on 57.8% of the targets in the DUDE benchmark.

Subfields: Computational Theory and Mathematics, Computer Vision and Pattern Recognition, Materials Chemistry, Molecular Biology, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Computer Vision and Pattern Recognition

In Materials Chemistry

In Molecular Biology

In Computational Chemistry & Drug Discovery

Citations in the corpus

Opportunities and obstacles for deep learning in biology and medicine (2017, background): “AtomNet [34] samples multiple ligand poses within the target binding site, and DeepVS [318] and Ragoza et al.”
Using Rule-Based Labels for Weak Supervised Learning: A ChemNet for Transferable Chemical Property Prediction (2017, methodology): “Other approaches use 2D or 3D images to train convolutional neural network (CNN) models [15, 16, 32], or SMILES strings to train recurrent neural network (RNN) models [3, 13].”
Tensor Field Networks: Rotation- and Translation-Equivariant Neural Networks for 3D Point Clouds (2018, methodology): “The other major approach to modeling 3D atomic systems is to voxelize the space (Wallach et al., 2015; Boomsma & Frellsen, 2017; Torng & Altman, 2017).”
EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction (2022, methodology): “…docked poses, as well as direct binding afﬁnity prediction from the 3D complex (McNutt et al., 2021; Fran-coeur et al., 2020; Ragoza et al., 2017; Wallach et al., 2015; Lim et al., 2019; Morrone et al., 2020; Jiang et al., 2021; Shen et al., 2021; Jastrzebski et al., 2020; Bao et al., 2021;…”
Deep convolutional networks for quality assessment of protein folds (2018, background): “native” or “not near native”. More in line with the “end-to-end” spirit of deep learning, methods using as input a 3D representation of the structure have been developed to score protein-ligand poses [16], [17], to predict ligandbinding protein pockets [18], and to predict the effect of a protein mu”
A Survey on Machine Learning Techniques for Cyber Security in the Last Decade (2020, methodology): “CNN is widely being used in image recognition [242], drug discovery [243], and anomaly detection [244], [245], to name a few.”