Filters








551 Hits in 5.4 sec

Protein Docking Model Evaluation by 3D Deep Convolutional Neural Networks

Xiao Wang, Genki Terashi, Charles W Christoffer, Mengmeng Zhu, Daisuke Kihara, Yann Ponty
2019 Bioinformatics  
Results We developed a convolutional deep neural network-based approach named DOVE (DOcking decoy selection with Voxel-based deep neural nEtwork) for evaluating protein docking models.  ...  To evaluate a protein docking model, DOVE scans the protein-protein interface of the model with a 3D voxel and considers atomic interaction types and their energetic contributions as input features applied  ...  Our method, DOcking decoy selection with Voxel-based deep neural nEtwork (DOVE), takes a docking decoy structure as input, maps the structure into a 3D grid, scans the protein-protein interface with a  ... 
doi:10.1093/bioinformatics/btz870 pmid:31746961 pmcid:PMC7141855 fatcat:kuzpketnj5daxl7rmnpczsfsre

Development of a graph convolutional neural network model for efficient prediction of protein-ligand binding affinities

Jeongtae Son, Dongsup Kim, Narcis Fernandez-Fuentes
2021 PLoS ONE  
We propose a novel deep-learning-based prediction model based on a graph convolutional neural network, named GraphBAR, for protein-ligand binding affinity.  ...  Graph convolutional neural networks reduce the computational time and resources that are normally required by the traditional convolutional neural network models.  ...  Pafnucy [2] is another convolutional neural network model that predicts protein-ligand binding affinities by using 3D CNNs.  ... 
doi:10.1371/journal.pone.0249404 pmid:33831016 fatcat:e5vvdu3nsnb3vp5gwgkskefv4i

MCN-CPI: Multiscale Convolutional Network for Compound–Protein Interaction Prediction

Shuang Wang, Mingjian Jiang, Shugang Zhang, Xiaofeng Wang, Qing Yuan, Zhiqiang Wei, Zhen Li
2021 Biomolecules  
However, the compound–protein interaction is complicated and the features extracted by most deep models are not comprehensive, which limits the performance to a certain extent.  ...  In this paper, we proposed a multiscale convolutional network that extracted the local and global features of the protein and the topological feature of the compound using different types of convolutional  ...  In the proposed compound-protein interaction predictive model, there are three convolutional networks, including the 3D convolutional network performed on the 3D structure of the binding site, the 1D convolutional  ... 
doi:10.3390/biom11081119 pmid:34439785 pmcid:PMC8392217 fatcat:csoiobl5aralxflupgw4445nkq

Combining docking pose rank and structure with deep learning improves protein-ligand binding mode prediction [article]

Joseph A. Morrone, Jeffrey K. Weber, Tien Huynh, Heng Luo, Wendy D. Cornell
2019 arXiv   pre-print
We present a simple, modular graph-based convolutional neural network that takes structural information from protein-ligand complexes as input to generate models for activity and binding mode prediction  ...  We develop a deep learning model for binding mode prediction that uses docking ranking as input in combination with docking structures.  ...  Each graph is processed by a convolutional neural network.  ... 
arXiv:1910.02845v1 fatcat:sw7czmx4mzhnfpmigxdnedvsxu

ResAtom System: Protein and Ligand Affinity Prediction Model Based on Deep Learning [article]

Yeji Wang, Shuo Wu, Yanwen Duan, Yong Huang
2021 arXiv   pre-print
Results: We build a predictive model of protein-ligand affinity through the ResNet neural network with added attention mechanism.  ...  The existing affinity prediction and evaluation functions based on deep learning mostly rely on experimentally-determined conformations.  ...  This work was also supported in part by the High Performance Computing Center of Central South University.  ... 
arXiv:2105.05125v1 fatcat:tcyp7xqwwfcx7bsknvo4eujdga

DeepRank: A deep learning framework for data mining 3D protein-protein interfaces [article]

Nicolas Renaud, Cunliang Geng, Sonja Georgievska, Francesco Ambrosetti, Lars Ridder, Dario F. Marzella, Alexandre M.J.J. Bonvin, Li C. Xue
2021 bioRxiv   pre-print
We present here DeepRank, a general, configurable deep learning framework for data mining PPIs using 3D convolutional neural networks (CNNs).  ...  The vast amount of experimentally and computationally resolved protein-protein interfaces (PPIs) offers the possibility of training deep learning models to aid the predictions of their biological relevance  ...  Wang et al. 5 have trained 3D deep convolutional networks (CNNs) on 3D grids representing protein-protein interfaces to evaluate the quality of docking models (DOVE 5 ).  ... 
doi:10.1101/2021.01.29.425727 fatcat:nubuph4jlveftohrx3v2c2q3ci

A Deep-Learning Approach toward Rational Molecular Docking Protocol Selection

José Jiménez-Luna, Alberto Cuzzolin, Giovanni Bolcato, Mattia Sturlese, Stefano Moro
2020 Molecules  
In this study, we developed a machine-learning model that uses a combination of convolutional and fully connected neural networks for the task of predicting the performance of several popular docking protocols  ...  We also rigorously evaluated the performance of our model using a widely available database of protein-ligand complexes and different types of data splits.  ...  A fully connected neural network handles ECFP4 fingerprints and descriptors computed from RDKit while a 3D-convolutional neural network processes a voxelized representation of the protein binding site.  ... 
doi:10.3390/molecules25112487 pmid:32471211 pmcid:PMC7321124 fatcat:jtzkyz5lpnb6fcjyhtb34pm6uy

Learning physics confers pose-sensitivity in structure-based virtual screening [article]

Pawel Gniewek and Bradley Worley and Kate Stafford and Henry van den Bedem and Brandon Anderson
2021 arXiv   pre-print
The conditioning forces the model to learn details of physical interactions. We evaluate these models on a new benchmark designed to detect pose-sensitivity.  ...  Here, we overcome this limitation by introducing a class of models with two key features: 1) we condition bioactivity on pose quality score, and 2) we present poor poses of true binders to the model as  ...  Table 1 : 1 Deep Learning Neural Networks used in this work. Architectures are either a 3D CNNbased model or a GCN-based model.  ... 
arXiv:2110.15459v3 fatcat:5ugoyrwiefdjfmfvxd3bpnxkki

DeepInterface: Protein-protein interface validation using 3D Convolutional Neural Networks [article]

Ali Tugrul Balci, Can Gumeli, Asma Hakouz, Deniz Yuret, Ozlem Keskin, Attila Gursoy
2019 bioRxiv   pre-print
The model is a 3-dimensional convolutional neural networks model and the positive datasets are obtained from all complexes in the Protein Data Bank, the negative datasets are the incorrect solutions of  ...  the docking decoys.  ...  DeepInterface Convolutional Neural Network Our network is composed of 4 convolutional layers followed by a global pooling layer and 2 fully connected layers.  ... 
doi:10.1101/617506 fatcat:lr6wsm3dorayfh6qnluoukt544

OnionNet: a Multiple-Layer Intermolecular-Contact-Based Convolutional Neural Network for Protein–Ligand Binding Affinity Prediction

Liangzhen Zheng, Jingrong Fan, Yuguang Mu
2019 ACS Omega  
In this study, a deep convolutional neural network model (called OnionNet) is introduced; its features are based on rotation-free element-pair-specific contacts between ligands and protein atoms, and the  ...  The robustness of the model is further explored by predicting the binding affinities of the complexes generated from docking simulations instead of experimentally determined PDB structures.  ...  [10] [11] [12] [13] 17 Taking the PDBBind v2016 core set database as the testing set, our OnionNet model obtained very similar performance to K DEEP, 17 a latest 3D-convolutional neural network model  ... 
doi:10.1021/acsomega.9b01997 pmid:31592466 pmcid:PMC6776976 fatcat:3h7jyabqa5gfxphgivbi3iapem

AttentionSiteDTI: Attention Based Model for Predicting Drug-Target Interaction Using 3D Structure of Protein Binding Sites [article]

Mehdi Yazdani-Jahromi, Niloofar Yousefi, Aida Tayebi, Ozlem Ozmen Garibay, Sudipta Seal, Elayaraja Kolanthai, Craig Neal
2021 bioRxiv   pre-print
Our proposed model utilize the binding sites (pockets) of the proteins as the input for the target protein, and it uses a self-attention mechanism to make the model learn which binding sites of the protein  ...  In addition, through multidisciplinary collaboration in this work, we further experimentally evaluate the practical potential of our proposed approach.  ...  has used the 3D structure of the protein and the structure-based, deep convolutional neural network to predict the binding of drug-target pairs.  ... 
doi:10.1101/2021.12.07.471693 fatcat:gulukrznljf2vn2kcqlzwg2apy

Artificial Intelligence in Aptamer–Target Binding Prediction

Zihao Chen, Long Hu, Bao-Ting Zhang, Aiping Lu, Yaofeng Wang, Yuanyuan Yu, Ge Zhang
2021 International Journal of Molecular Sciences  
Therefore, perspectives for models, algorithms, and implementation strategies of machine/deep learning-based methods are discussed.  ...  On the other hand, advanced machine-/deep-learning models have witnessed successes in predicting the binding abilities between targets and ligands in drug discovery and thus potentially offer a robust  ...  convolutional neural networks (CNN).  ... 
doi:10.3390/ijms22073605 pmid:33808496 fatcat:efmv3lbfkbhktopagqdtutkkem

Improved Protein-ligand Binding Affinity Prediction with Structure-Based Deep Fusion Inference [article]

Derek Jones, Hyojin Kim, Xiaohua Zhang, Adam Zemla, Garrett Stevenson, William D. Bennett, Dan Kirshner, Sergio Wong, Felice Lightstone, Jonathan E. Allen
2020 arXiv   pre-print
Despite the recent advances in the deep convolutional and graph neural network based approaches, the model performance depends on the input data representation and suffers from distinct limitations.  ...  We present fusion models to benefit from different feature representations of two neural network models to improve the binding affinity prediction.  ...  Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-JRNL-804162.  ... 
arXiv:2005.07704v1 fatcat:tpr6aowppzapdam5gbz5azbmwa

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

Izhar Wallach and Michael Dzamba and Abraham Heifets
2015 arXiv   pre-print
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.  ...  Structure-based deep-convolutional neural network The network topology consists of an input layer, followed by multiple 3D-convolutional and fullyconnected layers, and topped by a logistic-cost layer that  ... 
arXiv:1510.02855v1 fatcat:5u33t57gjjfihe5fjtzqekv2ei

NeuralDock: Rapid and Conformation-Agnostic Docking of Small Molecules

Congzhou M. Sha, Jian Wang, Nikolay V. Dokholyan
2022 Frontiers in Molecular Biosciences  
Here, we propose a neural network framework, NeuralDock, which accelerates the process of high-quality computational docking by a factor of 106, and does not require prior knowledge of a ligand that binds  ...  protein pocket 3D structure and small molecule topology.  ...  We train the neural network to directly predict the minimum binding energy evaluated by MedusaDock, based on a coarse 3D representation of the protein and a graph representation of the small molecule.  ... 
doi:10.3389/fmolb.2022.867241 pmid:35392534 pmcid:PMC8980736 fatcat:3zd3lul6ubhxhjergwfx2lzoxq
« Previous Showing results 1 — 15 out of 551 results