Protein-Ligand Scoring with Convolutional Neural Networks

By Matthew Ragoza, Joshua E. Hochuli, Elisa Idrobo, Jocelyn Sunseri, David Ryan Koes. 2016

313 citations recorded by the source corpus.

Computational approaches to drug discovery can reduce the time and cost associated with experimental assays and enable the screening of novel chemotypes. Structure-based drug design methods rely on scoring functions to rank and predict binding affinities and poses. The ever-expanding amount of protein-ligand binding and structural data enables the use of deep machine learning techniques for protein-ligand scoring. We describe convolutional neural network (CNN) scoring functions that take as input a comprehensive three-dimensional (3D) representation of a protein-ligand interaction. A CNN scoring function automatically learns the key features of protein-ligand interactions that correlate with binding. We train and optimize our CNN scoring functions to discriminate between correct and incorrect binding poses and known binders and nonbinders. We find that our CNN scoring function outperforms the AutoDock Vina scoring function when ranking poses both for pose prediction and virtual screening.

Subfields: Computational Theory and Mathematics, Molecular Biology

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 Molecular Biology

Citations in the corpus

PotentialNet for Molecular Property Prediction (2018, methodology): “More recent approaches include RF-Score 24,25 , NN-score 26 , Grid Featurizer 3 , three dimensional CNN approaches 27,28 , and Atomic Convolutional Neural Networks 29 .”
Machine Learning Methods in Drug Discovery (2020, methodology): “In the process of drug discovery, CNNs have become e ﬃ cient tools used in target discovery, lead discovery and characterization, in silico target-lead interaction screening, and protein-ligand scoring [68,88–90].”
Automating drug discovery (2017, background): “Several heterogeneous deep-learning systems with high prediction accuracies have been developed for drug–target association, drug repurposing opportunities and target identification, among other tasks 202,206,207,208 . and software 180,181 .”
The rise of deep learning in drug discovery. (2018, methodology): “A typical example is the investigation done by Ragoza et al. [80].”
Artificial intelligence in drug discovery: recent advances and future perspectives (2021, unknown): “Early approaches were inspired by the advances of computer vision and image recognition which were mostly driven by convolutional neural networks [76], and ultimately adapted for bioactivity prediction [77–80].”
ScanNet: an interpretable geometric deep learning model for structure-based protein binding site prediction (2021, background): “as sequences [22, 23], residue graphs [24–27], atomic density maps [28–34], atomic point clouds [35] or molecular surfaces [36, 37], each capturing different functionally relevant features.”
Compound‐protein interaction prediction with end‐to‐end learning of neural networks for graphs and sequences (2018, background): “(2015) and Ragoza et al. (2017) Compound–protein interaction prediction with end-to-end learning of GNN and CNN 315”
Key Topics in Molecular Docking for Drug Design (2019, methodology): “Deep learning neural networks have recently been applied to pose prediction and ranking [78,173,177,183,196].”
Artificial intelligence in cancer target identification and drug discovery (2022, background): “Currently, arti ﬁ cial intelligence biology analysis has bene-ﬁ ted from the utilization of graph-based neural networks instead of commonly used non-graph neural networks such as CNN 170 or DNN 169 , because graph-based neural networks can take the biological network structure as the input directly,”
Predicting drug–protein interaction using quasi-visual question answering system (2019, methodology): “, 2013]; 2) machine-learning scoring functions NN-Score [Durrant and McCammon, 2011]; RF-Score [Ballester and Mitchell, 2010]; 3) deep learning-based method 3D-CNN [Ragoza et al., 2017]; AtomNet [Wallach et al.”
DeepDTA: deep drug–target binding affinity prediction (2018, methodology): “…protein-ligand interaction binding affinities has been the focus of protein-ligand scoring, which is frequently used after virtual screening and docking campaigns in order to predict the putative strengths of the proposed ligands to the target for identifying the active and inactive compounds [43].”
Shape-Based Generative Modeling for de Novo Drug Design (2019, methodology): “Neural-network-inspired models have been recently used in computational biology 1 and chemistry to perform protein − ligand activity classi ﬁ cation, 2,3 a ﬃ nity 4 and property prediction, 5 design of synthetic routes, 6,7 and protein folding evaluation.”