8,902 Hits in 6.5 sec

Weighted protein residue networks based on joint recurrences between residues

Wael I. Karain, Nael I. Qaraeen
2015 BMC Bioinformatics  
Weighted and un-weighted protein residue networks can predict key functional residues in proteins based on the closeness centrality C and betweenness centrality B values for each residue.  ...  Results: The joint recurrence weighted network approach performs well in pointing out key protein residues.  ...  Weighted and un-weighted protein residue interaction networks are based mostly on a static protein structure. The network nodes represent either the carbon alpha or beta atoms.  ... 
doi:10.1186/s12859-015-0621-1 pmid:26003989 pmcid:PMC4491895 fatcat:xnub5tskfjhdvhkcitsidogbkm

Deep learning methods for protein torsion angle prediction

Haiou Li, Jie Hou, Badri Adhikari, Qiang Lyu, Jianlin Cheng
2017 BMC Bioinformatics  
The architectures including deep neural network (DNN) and deep restricted Boltzmann machine (DRBN), deep recurrent neural network (DRNN) and deep recurrent restricted Boltzmann machine (DReRBM) since the  ...  The deep recurrent network architecture performs slightly better than deep feed-forward architecture, and the predicted residue contact number and the error distribution of torsion angles extracted from  ...  The training of RBM is completely energy-guided based on the joint probability of all visible and hidden nodes, which is described by the following equation: E v; h ð Þ ¼ − P i b i v i − P j c j h j −  ... 
doi:10.1186/s12859-017-1834-2 pmid:28923002 pmcid:PMC5604354 fatcat:gucoqton7nh2xinej4r7hmxzve

Protein Solvent-Accessibility Prediction by a Stacked Deep Bidirectional Recurrent Neural Network

Buzhong Zhang, Linqing Li, Qiang Lü
2018 Biomolecules  
In this work, we present a deep learning method to predict residue solvent accessibility, which is based on a stacked deep bidirectional recurrent neural network applied to sequence profiles.  ...  Sequence-derived features including position-specific scoring matrix, physical properties, physicochemical characteristics, conservation score and protein coding were used to represent a residue.  ...  We focused on a RNN model specific to protein sequences and applied a bidirectional recurrent neural network (BRNN) to predict RSA.  ... 
doi:10.3390/biom8020033 pmid:29799510 pmcid:PMC6023031 fatcat:sfpel3ts3jeybd3egswqefv7yu

Empirical Study of Protein Feature Representation on Deep Belief Networks Trained with Small Data for Secondary Structure Prediction

Shamima Rashid, Suresh Sundaram, Chee Keong Kwoh
2022 IEEE/ACM Transactions on Computational Biology & Bioinformatics  
Here, the previous approach is extended to Deep Belief Networks (DBN).  ...  Testing on an independent set of G Switch proteins improved the Q3 score of the previous compact model by almost 3%.  ...  Recently, DNSS2 integrated layers of diverse network types including convolutional, residual, and recurrent layers among others and reported a Q 3 of 84.6% [28] .  ... 
doi:10.1109/tcbb.2022.3168676 pmid:35439138 fatcat:4nkjp3y5fbh23c2fjtezlhu6rm

Structure-based network analysis of an evolved G protein-coupled receptor homodimer interface

Sara E. Nichols, Carlos X. Hernández, Yi Wang, James Andrew McCammon
2013 Protein Science  
By applying principles from network analysis, sequence-based approaches such as statistical coupling analysis to determine coevolutionary residues, can be used in conjunction with molecular dynamics simulations  ...  Furthermore, this method may be applied to any protein-protein interaction.  ...  Structure-based network analysis based on coevolutionary relationship.  ... 
doi:10.1002/pro.2258 pmid:23553730 pmcid:PMC3690714 fatcat:wp4mtrgwwjhrza5zh6yve7zc5i

Interpretable Structured Learning with Sparse Gated Sequence Encoder for Protein-Protein Interaction Prediction [article]

Kishan KC, Feng Cui, Anne Haake, Rui Li
2020 arXiv   pre-print
We further employ a sparse regularization to model long-range dependencies between amino acids and to select important amino acids (protein motifs), thus enhancing interpretability.  ...  Literature-based case studies illustrate the ability of our model to provide biological insights to interpret the predictions.  ...  They use encoder based on a convolutional neural network (CNN) to capture local features and recurrent neural network (RNN) to capture sequential and contextualized features from protein sequences.  ... 
arXiv:2010.08514v1 fatcat:npycv5jm6bcyvmpffxzjxippwq

Deep Recurrent Conditional Random Field Network for Protein Secondary Prediction

Alexander Rosenberg Johansen, Casper Kaae Sønderby, Søren Kaae Sønderby, Ole Winther
2017 Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology,and Health Informatics - ACM-BCB '17  
Deep learning has become the state-of-the-art method for predicting protein secondary structure from only its amino acid residues and sequence pro le.  ...  We condition the CRF on the output of biRNN, which learns a distributed representation based on the entire sequence.  ...  Bi-directional Recurrent Neural Network A recurrent neural network (RNN) is a type of neural network layer that repeatedly uses the same weights along a sequence of data with a prior that the sequence  ... 
doi:10.1145/3107411.3107489 dblp:conf/bcb/JohansenSSW17 fatcat:eeilcpovdrebxnxja7xgjt7o3e

Cross-Modality Protein Embedding for Compound-Protein Affinity and Contact Prediction [article]

Yuning You, Yang Shen
2020 bioRxiv   pre-print
In this study we consider proteins as multi-modal data including 1D amino-acid sequences and (sequence-predicted) 2D residue-pair contact maps.  ...  We further propose two models involving cross-modality protein embedding and establish that the one with cross interaction (thus capturing correlations among modalities) outperforms SOTAs and our single  ...  Graph neural network (GNN, [14, 15, 16, 17, 18, 19] ) is adopted for compound 2D chemical graphs and hierarchical recurrent neural network (HRNN, [20] ) is chosen for protein 1D amino-acid sequences.  ... 
doi:10.1101/2020.11.29.403162 fatcat:hg26364lrrbp5cavny3bvzpqnu

Divide and Conquer Strategies for Protein Structure Prediction [chapter]

Pietro Di Lena, Piero Fariselli, Luciano Margara, Marco Vassura, Rita Casadio
2010 Mathematical Approaches to Polymer Sequence Analysis and Related Problems  
These two protein structure sub-problems are discussed in the light of the current evaluation of the performance that are based on periodical blind-checks (CASP meetings) and permanent evaluation (EVA  ...  As a second example, we also discuss the problem of predicting residue-residue contacts in proteins.  ...  The exceptions are PORTER, SCRATCH, SSPro4 (based on bidirectional recurrent NN), SAM-T99sec (based on HMM) and Yaspin (based both on NN and HMM).  ... 
doi:10.1007/978-1-4419-6800-5_2 fatcat:cfitt34ltvcbzagbfiqidadyf4

End-to-End Differentiable Learning of Protein Structure

Mohammed AlQuraishi
2018 Social Science Research Network  
Despite significant progress made by co-evolution methods to predict protein structure from signatures of residue-residue coupling found in the evolutionary record, a direct and explicit mapping between  ...  to its global representation via recurrent geometric units, and (4) the use of a differentiable loss function to capture deviations between predicted and experimental structures.  ...  While many metrics exist for FIGURE 2: Overview of Recurrent Geometric Networks.  ... 
doi:10.2139/ssrn.3239970 fatcat:gwwufc6dtjcydkkdrnedehxft4

Prediction of 8-state protein secondary structures by a novel deep learning architecture

Buzhong Zhang, Jinyan Li, Qiang Lü
2018 BMC Bioinformatics  
Furthermore, the residual network can improve the information flow between the hidden layers and the cascaded recurrent neural network.  ...  Results: We present a novel deep learning architecture which exploits an integrative synergy of prediction by a convolutional neural network, residual network, and bidirectional recurrent neural network  ...  Funding This work is supported in part by National Natural Science Foundation of China (No.61170125), the Special Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund (the  ... 
doi:10.1186/s12859-018-2280-5 pmid:30075707 pmcid:PMC6090794 fatcat:hcp5lntr6nh3hchvcjbm2b5jju

Learning to Evolve Structural Ensembles of Unfolded and Disordered Proteins Using Experimental Solution Data [article]

Oufan Zhang, Mojtaba Haghighatlari, Jie Li, Joao Miguel Correia Teixeira, Ashley Namini, Zi-Hao Liu, Julie D Forman-Kay, Teresa Head-Gordon
2022 arXiv   pre-print
We have developed a Generative Recurrent Neural Networks (GRNN) that learns the probability of the next residue torsions X_i+1=[ϕ_i+1,ψ_i+1,ω _i+1, χ_i+1] from the previous residue in the sequence X_i  ...  We show that updating the generative model parameters according to the reward feedback on the basis of the agreement between structures and data improves upon existing approaches that simply reweight static  ...  The generative model contains 2 recurrent units, one for recursion between residues and one for recursion between torsion angles within a residue.  ... 
arXiv:2206.12667v3 fatcat:mba45f5pirhlfpnvaaacom5rr4

A Novel Prediction Method for ATP-binding Sites from Protein Primary Sequences Based on Fusion of Deep Convolutional Neural Network and Ensemble Learning

Jiazhi Song, Yanchun Liang, Guixia Liu, Rongquan Wang, Liyan Sun, Ping Zhang
2020 IEEE Access  
Two classification networks are developed including a residual-inception-based predictor and a multi-inception-based predictor, then the ensemble learning is applied by giving optimized weights to each  ...  In this work, we introduce the deep convolutional neural network for protein-ATP binding sites prediction based on sequence information.  ...  Hassanzadeh and Wang [37] developed DeeperBind for sequence specificities of DNA binding proteins prediction with the combination of convolutional neural network and recurrent neural network.  ... 
doi:10.1109/access.2020.2968847 fatcat:4hlrw27zevcmrnbdylffzlsvk4

Protein Design with Deep Learning

Marianne Defresne, Sophie Barbe, Thomas Schiex
2021 International Journal of Molecular Sciences  
Computational Protein Design (CPD) has produced impressive results for engineering new proteins, resulting in a wide variety of applications.  ...  In the case of protein data, specific representations are needed for both the amino acid sequence and the protein structure in order to capture respectively 1D and 3D information.  ...  on the weights (for training), but also on the input.  ... 
doi:10.3390/ijms222111741 pmid:34769173 pmcid:PMC8584038 fatcat:rd7elqaiefgvrp3rvemjyaj3ma

Explainable Deep Relational Networks for Predicting Compound-Protein Affinities and Contacts [article]

Mostafa Karimi, Di Wu, Zhangyang Wang, Yang Shen
2019 arXiv   pre-print
Predicting compound-protein affinity is critical for accelerating drug discovery. Recent progress made by machine learning focuses on accuracy but leaves much to be desired for interpretability.  ...  We further design a physics-inspired deep relational network, DeepRelations, with intrinsically explainable architecture.  ...  In the last stage, T [3] ij = W [2] ij focuses on putative contacts between protein residues and compound atoms.  ... 
arXiv:1912.12553v1 fatcat:sslcp6vt5nejnortd47aqzkopi
« Previous Showing results 1 — 15 out of 8,902 results