scholarly journals SAINT: Self-Attention Augmented Inception-Inside-Inception Network Improves Protein Secondary Structure Prediction

2019 ◽  
Author(s):  
Mostofa Rafid Uddin ◽  
Sazan Mahbub ◽  
M Saifur Rahman ◽  
Md Shamsuzzoha Bayzid
Keyword(s):  
Secondary Structure ◽  
Language Processing ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Accurate Method ◽  
Attention Mechanism ◽  
Resonance Spectroscopy ◽  
Major Step ◽  
Extensive Evaluation

AbstractMotivationProtein structures provide basic insight into how they can interact with other proteins, their functions and biological roles in an organism. Experimental methods (e.g., X-ray crystallography, nuclear magnetic resonance spectroscopy) for predicting the secondary structure (SS) of proteins are very expensive and time consuming. Therefore, developing efficient computational approaches for predicting the secondary structure of protein is of utmost importance. Advances in developing highly accurate SS prediction methods have mostly been focused on 3-class (Q3) structure prediction. However, 8-class (Q8) resolution of secondary structure contains more useful information and is much more challenging than the Q3 prediction.ResultsWe present SAINT, a highly accurate method for Q8 structure prediction, which incorporates self-attention mechanism (a concept from natural language processing) with the Deep Inception-Inside-Inception (Deep3I) network in order to effectively capture both the short-range and long-range interactions among the amino acid residues. SAINT offers a more interpretable framework than the typical black-box deep neural network methods. Through an extensive evaluation study, we report the performance of SAINT in comparison with the existing best methods on a collection of benchmark datasets, namely, TEST2016, TEST2018, CASP12 and CASP13. Our results suggest that self-attention mechanism improves the prediction accuracy and outperforms the existing best alternate methods. SAINT is the first of its kind and offers the best known Q8 accuracy. Thus, we believe SAINT represents a major step towards the accurate and reliable prediction of secondary structures of proteins.AvailabilitySAINT is freely available as an open source project at https://github.com/SAINTProtein/SAINT.

scholarly journals SAINT: self-attention augmented inception-inside-inception network improves protein secondary structure prediction

2020 ◽  
Vol 36 (17) ◽  
pp. 4599-4608 ◽  
Author(s):  
Mostofa Rafid Uddin ◽  
Sazan Mahbub ◽  
M Saifur Rahman ◽  
Md Shamsuzzoha Bayzid
Keyword(s):  
Secondary Structure ◽  
Language Processing ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Accurate Method ◽  
Attention Mechanism ◽  
Resonance Spectroscopy ◽  
Major Step ◽  
Extensive Evaluation

Abstract Motivation Protein structures provide basic insight into how they can interact with other proteins, their functions and biological roles in an organism. Experimental methods (e.g. X-ray crystallography and nuclear magnetic resonance spectroscopy) for predicting the secondary structure (SS) of proteins are very expensive and time consuming. Therefore, developing efficient computational approaches for predicting the SS of protein is of utmost importance. Advances in developing highly accurate SS prediction methods have mostly been focused on 3-class (Q3) structure prediction. However, 8-class (Q8) resolution of SS contains more useful information and is much more challenging than the Q3 prediction. Results We present SAINT, a highly accurate method for Q8 structure prediction, which incorporates self-attention mechanism (a concept from natural language processing) with the Deep Inception-Inside-Inception network in order to effectively capture both the short- and long-range interactions among the amino acid residues. SAINT offers a more interpretable framework than the typical black-box deep neural network methods. Through an extensive evaluation study, we report the performance of SAINT in comparison with the existing best methods on a collection of benchmark datasets, namely, TEST2016, TEST2018, CASP12 and CASP13. Our results suggest that self-attention mechanism improves the prediction accuracy and outperforms the existing best alternate methods. SAINT is the first of its kind and offers the best known Q8 accuracy. Thus, we believe SAINT represents a major step toward the accurate and reliable prediction of SSs of proteins. Availability and implementation SAINT is freely available as an open-source project at https://github.com/SAINTProtein/SAINT.

scholarly journals Hermes: an ensemble machine learning architecture for protein secondary structure prediction

2019 ◽  
Author(s):  
Larry Bliss ◽  
Ben Pascoe ◽  
Samuel K Sheppard
Keyword(s):  
Machine Learning ◽  
Protein Structure ◽  
Secondary Structure ◽  
Structure Prediction ◽  
Cross Validation ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Lower Boundary ◽  
Protein Secondary Structure ◽  
Homologous Proteins

AbstractMotivationProtein structure predictions, that combine theoretical chemistry and bioinformatics, are an increasingly important technique in biotechnology and biomedical research, for example in the design of novel enzymes and drugs. Here, we present a new ensemble bi-layered machine learning architecture, that directly builds on ten existing pipelines providing rapid, high accuracy, 3-State secondary structure prediction of proteins.ResultsAfter training on 1348 solved protein structures, we evaluated the model with four independent datasets: JPRED4 - compiled by the authors of the successful predictor with the same name, and CASP11, CASP12 & CASP13 - assembled by the Critical Assessment of protein Structure Prediction consortium who run biannual experiments focused on objective testing of predictors. These rigorous, pre-established protocols included 7-fold cross-validation and blind testing. This led to a mean Hermes accuracy of 95.5%, significantly (p<0.05) better than the ten previously published models analysed in this paper. Furthermore, Hermes yielded a reduction in standard deviation, lower boundary outliers, and reduced dependency on solved structures of homologous proteins, as measured by NEFF score. This architecture provides advantages over other pipelines, while remaining accessible to users at any level of bioinformatics experience.Availability and ImplementationThe source code for Hermes is freely available at: https://github.com/HermesPrediction/Hermes. This page also includes the cross-validation with corresponding models, and all training/testing data presented in this study with predictions and accuracy.

scholarly journals ProteinUnet2 for Fast Protein Secondary Structure Prediction: A Step Towards Proper Evaluation

2021 ◽  
Author(s):  
Katarzyna Stapor ◽  
Krzysztof Kotowski ◽  
Tomasz Smolarczyk ◽  
Irena Roterman
Keyword(s):  
Secondary Structure ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
State Of The Art ◽  
Protein Secondary Structure ◽  
Evaluation Study ◽  
Practical Significance ◽  
Statistical Methodology ◽  
Extensive Evaluation ◽  
Benchmark Datasets

Abstract Background: The importance of protein secondary structure (SS) prediction is widely known, its solution enables learning about the role of a protein in organisms. As the experimental methods are expensive and sometimes impossible, many SS predictors, mainly based on different machine learning methods have been proposed for many years. SS prediction as the imbalanced classification problem should not be judged by the commonly used Q3/Q8 metrics. Moreover, as the benchmark datasets are not random samples, the classical statistical null hypothesis testing based on the Neyman-Pearson approach is not appropriate. Also, the state-of-the-art predictors have usually relatively long prediction times.Results: We present a new deep network ProteinUnet2 for SS prediction which is based on U-Net convolutional architecture. We also propose a new statistical methodology for prediction performance assessment based on the significance from Fisher-Pitman permutation tests accompanied by practical significance measured by Cohen’s effect size. Through an extensive evaluation study, we report the performance of ProteinUnet2 in comparison with two state-of-the-art methods SAINT and SPOT-1D on benchmark datasets TEST2016, TEST2018, and CASP12. Conclusions: Our results suggest that ProteinUnet2 has much shorter prediction times while maintaining (or outperforming) the mentioned predictors. We strongly believe that our proposed statistical methodology will be adopted and used (and even expanded) by the research community.

scholarly journals Literature Survey of Protein Secondary Structure Prediction

2012 ◽  
Author(s):  
Satya Nanda Vel Arjunan ◽  
Safaai Deris ◽  
Rosli Md Illias
Keyword(s):  
Protein Structure ◽  
Secondary Structure ◽  
Structure Prediction ◽  
Large Scale ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Protein Secondary Structure ◽  
Fundamental Theory ◽  
Protein Secondary Structure Prediction ◽  
General Guide

Dengan wujudnya projek jujukan DNA secara besar-besaran, teknik yang tepat untuk meramalkan struktur protein diperlukan. Masalah meramalkan struktur protein daripada jujukan DNA pada dasarnya masih belum dapat diselesaikan walaupun kajian intensif telah dilakukan selama lebih daripada tiga dekad. Dalam kertas kerja ini, teori asas struktur protein akan dibincangkan sebagai panduan umum bagi kajian peramalan struktur protein sekunder. Analisis jujukan terkini serta prinsi p yang digunakan dalam teknik-teknik tersebut akan diterangkan. Kata kunci: peramalan stuktur sekunder protein; rangkaian neural. In the wake of large-scale DNA sequencing projects, accurate tools are needed to predict protein structures. The problem of predicting protein structure from DNA sequence remains fundamentally unsolved even after more than three decades of intensive research. In this paper, fundamental theory of the protein structure of the protein structure will be presented as a general guide to protein secondary structure prediction research. An overview of the state-of-theart in sequence analysis and some princi ples of the methods invloved wil be described. Key words: protein secondary structure prediction;neural networks.

Prediction of Protein Secondary Structure

2012 ◽  
Author(s):  
Satya Nanda Vel Arjunan ◽  
Safaai Deris ◽  
Rosli Md Illias
Keyword(s):  
Protein Structure ◽  
Secondary Structure ◽  
Structure Prediction ◽  
Large Scale ◽  
Secondary Structure Prediction ◽  
State Of The Art ◽  
Protein Structures ◽  
Protein Secondary Structure ◽  
Protein Secondary Structure Prediction ◽  
General Guide

Dengan wujudnya projek jujukan DNA secara besar–besaran, teknik yang tepat untuk meramalkan struktur protein diperlukan. Masalah meramalkan struktur protein daripada jujukan DNA pada dasarnya masih belum dapat diselesaikan walaupun kajian intensif telah dilakukan selama lebih daripada tiga dekad. Dalam kertas kerja ini, teori asas struktur protein akan dibincangkan sebagai panduan umum bagi kajian peramalan struktur protein sekunder. Analisis jujukan terkini serta prinsip yang digunakan dalam teknik–teknik tersebut akan diterangkan. Kata kunci: Peramalan struktur sekunder protein; Rangkaian Neural In the wake of large-scale DNA sequencing projects, accurate tools are needed to predict protein structures. The problem of predicting protein structure from DNA sequence remains fundamentally unsolved even after more than three decades of intensive research. In this paper, fundamental theory of the protein structure will be presented as a general guide to protein secondary structure prediction research. An overview of the state–of–the–art in sequence analysis and some principles of the methods involved wil be described. Key words: Protein secondary structure prediction; Neural networks

scholarly journals Improving protein structure prediction by deep learning and computational optimization

2019 ◽  
Author(s):  
◽  
Jie Hou
Keyword(s):  
Deep Learning ◽  
Protein Structure ◽  
Secondary Structure ◽  
Protein Structure Prediction ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Protein Secondary Structure ◽  
Scattering Data ◽  
Protein Secondary Structure Prediction

Protein structure prediction is one of the most important scientific problems in the field of bioinformatics and computational biology. The availability of protein three-dimensional (3D) structure is crucial for studying biological and cellular functions of proteins. The importance of four major sub-problems in protein structure prediction have been clearly recognized. Those include, first, protein secondary structure prediction, second, protein fold recognition, third, protein quality assessment, and fourth, multi-domain assembly. In recent years, deep learning techniques have proved to be a highly effective machine learning method, which has brought revolutionary advances in computer vision, speech recognition and bioinformatics. In this dissertation, five contributions are described. First, DNSS2, a method for protein secondary structure prediction using one-dimensional deep convolution network. Second, DeepSF, a method of applying deep convolutional network to classify protein sequence into one of thousands known folds. Third, CNNQA and DeepRank, two deep neural network approaches to systematically evaluate the quality of predicted protein structures and select the most accurate model as the final protein structure prediction. Fourth, MULTICOM, a protein structure prediction system empowered by deep learning and protein contact prediction. Finally, SAXSDOM, a data-assisted method for protein domain assembly using small-angle X-ray scattering data. All the methods are available as software tools or web servers which are freely available to the scientific community.

scholarly journals Isolating SARS-CoV-2 Strains From Countries in the Same Meridian: Genome Evolutionary Analysis

10.2196/25995 ◽  
2021 ◽  
Vol 2 (1) ◽  
pp. e25995
Author(s):  
Emilio Mastriani ◽  
Alexey V Rakov ◽  
Shu-Lin Liu
Keyword(s):  
Secondary Structure ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Data Repository ◽  
Evolutionary Analysis ◽  
Wild Type ◽  
Uncharacterized Protein ◽  
Climate Conditions

Background COVID-19, caused by the novel SARS-CoV-2, is considered the most threatening respiratory infection in the world, with over 40 million people infected and over 0.934 million related deaths reported worldwide. It is speculated that epidemiological and clinical features of COVID-19 may differ across countries or continents. Genomic comparison of 48,635 SARS-CoV-2 genomes has shown that the average number of mutations per sample was 7.23, and most SARS-CoV-2 strains belong to one of 3 clades characterized by geographic and genomic specificity: Europe, Asia, and North America. Objective The aim of this study was to compare the genomes of SARS-CoV-2 strains isolated from Italy, Sweden, and Congo, that is, 3 different countries in the same meridian (longitude) but with different climate conditions, and from Brazil (as an outgroup country), to analyze similarities or differences in patterns of possible evolutionary pressure signatures in their genomes. Methods We obtained data from the Global Initiative on Sharing All Influenza Data repository by sampling all genomes available on that date. Using HyPhy, we achieved the recombination analysis by genetic algorithm recombination detection method, trimming, removal of the stop codons, and phylogenetic tree and mixed effects model of evolution analyses. We also performed secondary structure prediction analysis for both sequences (mutated and wild-type) and “disorder” and “transmembrane” analyses of the protein. We analyzed both protein structures with an ab initio approach to predict their ontologies and 3D structures. Results Evolutionary analysis revealed that codon 9628 is under episodic selective pressure for all SARS-CoV-2 strains isolated from the 4 countries, suggesting it is a key site for virus evolution. Codon 9628 encodes the P0DTD3 (Y14_SARS2) uncharacterized protein 14. Further investigation showed that the codon mutation was responsible for helical modification in the secondary structure. The codon was positioned in the more ordered region of the gene (41-59) and near to the area acting as the transmembrane (54-67), suggesting its involvement in the attachment phase of the virus. The predicted protein structures of both wild-type and mutated P0DTD3 confirmed the importance of the codon to define the protein structure. Moreover, ontological analysis of the protein emphasized that the mutation enhances the binding probability. Conclusions Our results suggest that RNA secondary structure may be affected and, consequently, the protein product changes T (threonine) to G (glycine) in position 50 of the protein. This position is located close to the predicted transmembrane region. Mutation analysis revealed that the change from G (glycine) to D (aspartic acid) may confer a new function to the protein—binding activity, which in turn may be responsible for attaching the virus to human eukaryotic cells. These findings can help design in vitro experiments and possibly facilitate a vaccine design and successful antiviral strategies.

scholarly journals By using MADALINE Learning with Back Propagation and Keras to Predict the Protein Secondary Structure

2019 ◽  
Vol 9 (2) ◽  
pp. 4878-4882
Keyword(s):  
Protein Structure ◽  
Secondary Structure ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Back Propagation ◽  
Protein Secondary Structure ◽  
Sigmoid Function ◽  
Crucial Component ◽  
Main Motive

Understanding of intermediate protein structure prediction serves as a crucial component to find the function of residues of amino acid. In this paper, focus on the intermediate protein structure by using feed forward and feedback method and enhancing the concept of sliding window. Prediction of secondary structure is a very cosmic problem of bioinformatics. This can be reduced by predicting or unfold the protein structures if it is unfolded so that can give the great results in medical sciences. Our main motive is to improve the accuracy of secondary structures and minimize the error .Experimentally, use the Multilayer ADALINE network for learning and KERAS TENSORFLOW use for train the weight matrix and sigmoid function for calculating the resultant with back propagation. Resultant of this paper results provides more prominent results as compare to already existing methods. Those improve the accuracy of secondary structure prediction

scholarly journals The Future of Protein Secondary Structure Prediction Was Invented by Oleg Ptitsyn

Biomolecules ◽  
2020 ◽  
Vol 10 (6) ◽  
pp. 910
Author(s):  
Daniel Rademaker ◽  
Jarek van Dijk ◽  
Willem Titulaer ◽  
Joanna Lange ◽  
Gert Vriend ◽  
...  
Keyword(s):  
Secondary Structure ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Structures ◽  
Protein Secondary Structure ◽  
Prediction Method ◽  
Research Field ◽  
Protein Secondary Structure Prediction ◽  
Learning Methods ◽  
Average Percentage

When Oleg Ptitsyn and his group published the first secondary structure prediction for a protein sequence, they started a research field that is still active today. Oleg Ptitsyn combined fundamental rules of physics with human understanding of protein structures. Most followers in this field, however, use machine learning methods and aim at the highest (average) percentage correctly predicted residues in a set of proteins that were not used to train the prediction method. We show that one single method is unlikely to predict the secondary structure of all protein sequences, with the exception, perhaps, of future deep learning methods based on very large neural networks, and we suggest that some concepts pioneered by Oleg Ptitsyn and his group in the 70s of the previous century likely are today’s best way forward in the protein secondary structure prediction field.

scholarly journals Protein Secondary Structure Prediction With a Reductive Deep Learning Method

2021 ◽  
Vol 9 ◽  
Author(s):  
Zhiliang Lyu ◽  
Zhijin Wang ◽  
Fangfang Luo ◽  
Jianwei Shuai ◽  
Yandong Huang
Keyword(s):  
Deep Learning ◽  
Secondary Structure ◽  
Structure Prediction ◽  
Secondary Structure Prediction ◽  
Protein Secondary Structure ◽  
Secondary Structures ◽  
Resonance Spectroscopy ◽  
Data Set ◽  
X Ray Crystallography ◽  
Protein Secondary Structures

Protein secondary structures have been identified as the links in the physical processes of primary sequences, typically random coils, folding into functional tertiary structures that enable proteins to involve a variety of biological events in life science. Therefore, an efficient protein secondary structure predictor is of importance especially when the structure of an amino acid sequence fragment is not solved by high-resolution experiments, such as X-ray crystallography, cryo-electron microscopy, and nuclear magnetic resonance spectroscopy, which are usually time consuming and expensive. In this paper, a reductive deep learning model MLPRNN has been proposed to predict either 3-state or 8-state protein secondary structures. The prediction accuracy by the MLPRNN on the publicly available benchmark CB513 data set is comparable with those by other state-of-the-art models. More importantly, taking into account the reductive architecture, MLPRNN could be a baseline for future developments.

Sign in / Sign up

Export Citation Format

Share Document