Protein 3D structure-based neural networks highly improve the accuracy
in compound-protein binding affinity prediction
- URL: http://arxiv.org/abs/2204.12586v1
- Date: Wed, 30 Mar 2022 00:44:15 GMT
- Title: Protein 3D structure-based neural networks highly improve the accuracy
in compound-protein binding affinity prediction
- Authors: Binjie Guo, Hanyu Zheng, Huan Huang, Haohan Jiang, Xiaodan Li, Naiyu
Guan, Yanming Zuo, Yicheng Zhang, Hengfu Yang, Xuhua Wang
- Abstract summary: We develop Fast Evolutional Attention and Thoroughgoing-graph Neural Networks (FeatNN) to facilitate the application of protein 3D structure information for predicting compound-protein binding affinities (CPAs)
FeatNN considerably outperforms various state-of-the-art baselines in CPA prediction with the Pearson value elevated by about 35.7%.
- Score: 7.059949221160259
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Theoretically, the accuracy of computational models in predicting
compound-protein binding affinities (CPAs) could be improved by the
introduction of protein 3D structure information. However, most of these models
still suffer from a low accuracy due to the lack of an efficient approach to
encode informative protein features. The major challenge is how to combine the
multi-modal information such as the residue sequence of the protein, residue
atom coordinates and the torsion angles. To tackle this problem, we develop
Fast Evolutional Attention and Thoroughgoing-graph Neural Networks (FeatNN) to
facilitate the application of protein 3D structure information for predicting
CPAs. Specifically, we established a novel end-to-end architecture to jointly
embed torsion matrix, discrete distance matrix, and sequence information of
protein and extract compound features with deep graph convolution layers. In
addition, a new pairwise mapping attention mechanism is introduced to
comprehensively learn potential interaction information between proteins and
compounds. FeatNN considerably outperforms various state-of-the-art baselines
in CPA prediction with the Pearson value elevated by about 35.7%. Thus, FeatNN
provides an outstanding method for highly accurate CPA prediction and
facilitates high-throughput virtual screening of drug candidates.
Related papers
- Protein binding affinity prediction under multiple substitutions applying eGNNs on Residue and Atomic graphs combined with Language model information: eGRAL [1.840390797252648]
Deep learning is increasingly recognized as a powerful tool capable of bridging the gap between in-silico predictions and in-vitro observations.
We propose eGRAL, a novel graph neural network architecture designed for predicting binding affinity changes from amino acid substitutions in protein complexes.
eGRAL leverages residue, atomic and evolutionary scales, thanks to features extracted from protein large language models.
arXiv Detail & Related papers (2024-05-03T10:33:19Z) - DeepGATGO: A Hierarchical Pretraining-Based Graph-Attention Model for
Automatic Protein Function Prediction [4.608328575930055]
Automatic protein function prediction (AFP) is classified as a large-scale multi-label classification problem.
Currently, popular methods primarily combine protein-related information and Gene Ontology (GO) terms to generate final functional predictions.
We propose a sequence-based hierarchical prediction method, DeepGATGO, which processes protein sequences and GO term labels hierarchically.
arXiv Detail & Related papers (2023-07-24T07:01:32Z) - State-specific protein-ligand complex structure prediction with a
multi-scale deep generative model [68.28309982199902]
We present NeuralPLexer, a computational approach that can directly predict protein-ligand complex structures.
Our study suggests that a data-driven approach can capture the structural cooperativity between proteins and small molecules, showing promise in accelerating the design of enzymes, drug molecules, and beyond.
arXiv Detail & Related papers (2022-09-30T01:46:38Z) - DProQ: A Gated-Graph Transformer for Protein Complex Structure
Assessment [7.988932562855392]
DProQ is a gated neighborhood-modulating Graph Transformer (GGT) designed to predict the quality of 3D protein complex structures.
We incorporate node and edge gates within a novel Graph Transformer framework to control information flow during graph message passing.
Our rigorous experiments demonstrate that DProQ achieves state-of-the-art performance in ranking protein complex structures.
arXiv Detail & Related papers (2022-05-21T15:41:46Z) - Learning Geometrically Disentangled Representations of Protein Folding
Simulations [72.03095377508856]
This work focuses on learning a generative neural network on a structural ensemble of a drug-target protein.
Model tasks involve characterizing the distinct structural fluctuations of the protein bound to various drug molecules.
Results show that our geometric learning-based method enjoys both accuracy and efficiency for generating complex structural variations.
arXiv Detail & Related papers (2022-05-20T19:38:00Z) - Structure-aware Protein Self-supervised Learning [50.04673179816619]
We propose a novel structure-aware protein self-supervised learning method to capture structural information of proteins.
In particular, a well-designed graph neural network (GNN) model is pretrained to preserve the protein structural information.
We identify the relation between the sequential information in the protein language model and the structural information in the specially designed GNN model via a novel pseudo bi-level optimization scheme.
arXiv Detail & Related papers (2022-04-06T02:18:41Z) - Structure-aware Interactive Graph Neural Networks for the Prediction of
Protein-Ligand Binding Affinity [52.67037774136973]
Drug discovery often relies on the successful prediction of protein-ligand binding affinity.
Recent advances have shown great promise in applying graph neural networks (GNNs) for better affinity prediction by learning the representations of protein-ligand complexes.
We propose a structure-aware interactive graph neural network (SIGN) which consists of two components: polar-inspired graph attention layers (PGAL) and pairwise interactive pooling (PiPool)
arXiv Detail & Related papers (2021-07-21T03:34:09Z) - EBM-Fold: Fully-Differentiable Protein Folding Powered by Energy-based
Models [53.17320541056843]
We propose a fully-differentiable approach for protein structure optimization, guided by a data-driven generative network.
Our EBM-Fold approach can efficiently produce high-quality decoys, compared against traditional Rosetta-based structure optimization routines.
arXiv Detail & Related papers (2021-05-11T03:40:29Z) - PersGNN: Applying Topological Data Analysis and Geometric Deep Learning
to Structure-Based Protein Function Prediction [0.07340017786387766]
In this work, we isolate protein structure to make functional annotations for proteins in the Protein Data Bank.
We present PersGNN - an end-to-end trainable deep learning model that combines graph representation learning with topological data analysis.
arXiv Detail & Related papers (2020-10-30T02:24:35Z) - Transfer Learning for Protein Structure Classification at Low Resolution [124.5573289131546]
We show that it is possible to make accurate ($geq$80%) predictions of protein class and architecture from structures determined at low ($leq$3A) resolution.
We provide proof of concept for high-speed, low-cost protein structure classification at low resolution, and a basis for extension to prediction of function.
arXiv Detail & Related papers (2020-08-11T15:01:32Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.