Protein secondary structure prediction using modular reciprocal bidirectional recurrent neural networks (original) (raw)
Related papers
Protein Secondary Structure Prediction using Feed-Forward Neural Network
2010
Abstract—Neural network is one of the successful methods for protein secondary structure prediction. Day to day this technology is modified, improved, even other methods also combined with it to get better result. In this paper we trained feed-forward neural network with proteins for secondary structure prediction. Using Java Object Oriented Neural Engine (JOONE) our achieved accuracy for helix prediction is 71% and for sheet prediction is 65%.
The usage of Neural Networks Paradigm in the prediction of protein secondary structure
—The process of predicting the secondary structure of protein is crucial in understanding the protein functionality. It is very important in understanding the protein functions and in diagnosing any disorder. This paper shows the usage of neural networks paradigm in the prediction process. It discusses the difference between applying feed forward and back propagation algorithms on the prediction accuracy. The results show that the highest accuracy is reached when presenting the primary sequence in binary format and use Feed-Forward network. The accuracy reached around 86% when predicting Beta strands or sheets only. Moreover, it was seen that predicting alpha and beta each alone and combining the results didn't show significant accuracy enhancement.
Secondary structure predictions are increasingly becoming the workhorse for several methods aiming at predicting protein structure and function. Here we use ensembles of bidirectional recurrent neural network architectures, PSI-BLAST-derived profiles, and a large nonredundant training set to derive two new predictors: (a) the second version of the SSpro program for secondary structure classification into three categories and (b) the first version of the SSpro8 program for secondary structure classification into the eight classes produced by the DSSP program. We describe the results of three different test sets on which SSpro achieved a sustained performance of about 78% correct prediction. We report confusion matrices, compare PSI-BLAST to BLAST-derived profiles, and assess the corresponding performance improvements. SSpro and SSpro8 are implemented as web servers, available together with other structural feature predictors at: http://promoter.ics.uci.edu/ BRNN-PRED/. Proteins 2002;47:228 -235.
Knowing that predicting the secondary structure of protein can help deeply in the protein functionality understanding and helps in multiple diseases diagnoses. Moreover, it can help in giving an accurate prediction to the tertiary structures. This forces any researcher to try multiple experiments to find ways to enhance the prediction accuracy. Through the past decade, many machine learning techniques have been used to predict the secondary structures. The main used technique was neural networks. This paper presents three different experiments that make use of artificial neural networks. The first uses a single neural network with different number of hidden layers and nodes. The second combines the output of two neural networks to enhance the accuracy. Last, the third compared to the previous two predicts not only the broad classes of secondary structure (namely; alpha, beta and coil) but predicts all the classes, then combines the result. All the experiments are based on a data set pulled out from the Rcsb protein data bank. The results of all experiments show that the highest accuracy is reached when encoding the primary sequence using binary format and use Feed-Forward network. The accuracy reached around 86% when predicting Beta strands or sheets only. Combining the results of two neural networks showed accuracy of about 83%. Moreover, merging the results of alpha and beta predictors didn't show high significance.
Artificial Intelligence …, 2010
Successful protein secondary structure prediction is an important step towards modelling protein 3D structure, with several practical applications. Even though in the last four decades several PSSP algorithms have been proposed, we are far from being accurate. The Bidirectional Recurrent Neural Network (BRNN) architecture of Baldi et al. is currently considered as one of the optimal computational neural network type architectures for addressing the problem. In this paper, we implement the same BRNN architecture, but we use a modified training procedure. More specifically, our aim is to identify the effect of the contribution of local versus global information, by varying the length of the segment on which the Recurrent Neural Networks operate for each residue position considered. For training the network, the backpropagation learning algorithm with an online training procedure is used, where the weight updates occur for every amino acid, as opposed to Baldi et al.
Protein Structure Prediction using Artificial Neural Network
2011
Protein secondary structure prediction is a problem related to structural bioinformatics which deals with the prediction and analysis of macromolecules i.e. DNA, RNA and protein. It is an important step towards elucidating its three dimensional structure, as well as its function. Secondary structure of a protein can be predicted from its primary structures i.e. from the amino acid sequences or from the residues though challenges exists. For these four methods are used. These are Statistical Approach, Nearest Neighbor method, Neural Network Approach and Hidden Markov Model Approach. The Artificial Neural Network (ANN) approach for prediction of protein secondary structure is the most successful one among all the methods used. In this method, ANNs are trained to make them capable of performing recognition of amino acid patterns in known secondary structure units and these patterns are used to distinguish between the different types of secondary structures. This work is related to the prediction of secondary structure of proteins employing artificial neural network though it is restricted initially to three structures only.
Protein secondary structure prediction using neural networks and deep learning: A review
2019
Literature contains over fifty years of accumulated methods proposed by researchers for predicting the secondary structures of proteins in silico. A large part of this collection is comprised of artificial neural network-based approaches, a field of artificial intelligence and machine learning that is gaining increasing popularity in various application areas. The primary objective of this paper is to put together the summary of works that are important but sparse in time, to help new researchers have a clear view of the domain in a single place. An informative introduction to protein secondary structure and artificial neural networks is also included for context. This review will be valuable in designing future methods to improve protein secondary structure prediction accuracy. The various neural network methods found in this problem domain employ varying architectures and feature spaces, and a handful stand out due to significant improvements in prediction. Neural networks with la...
Reaching optimized parameter set: protein secondary structure prediction using neural network
Neural Computing and Applications, 2016
We propose an optimized parameter set for protein secondary structure prediction using three layer feed forward back propagation neural network. The methodology uses four parameters viz. encoding scheme, window size, number of neurons in the hidden layer and type of learning algorithm. The input layer of the network consists of neurons changing from 3 to 19, corresponding to different window sizes. The hidden layer chooses a natural number from 1 to 20 as the number of neurons. The output layer consists of three neurons, each corresponding to known secondary structural classes viz. α-helix, β-strands and coil/turns respectively. It also uses eight different learning algorithms and nine encoding schemes. Exhaustive experiments were performed using non-homologues dataset. The experimental results were compared using performance measures like Q 3 , sensitivity, specificity, Mathew correlation coefficient and accuracy. The paper also discusses the process of obtaining a stabilized cluster of 2530 records from a collection of 11340 records. The graphs of these stabilized clusters of records with respect to accuracy are concave, convergence is monotonic increasing and rate of convergence is uniform. The paper gives BLOSUM62 as the encoding scheme, 19 as the window size, 19 as the number of neurons in the hidden layer and One-Step Secant as the learning algorithm with the highest accuracy of 78%. These parameter values are proposed as the optimized parameter set for the three layer feed forward back propagation neural network for the protein secondary structure prediction.
Bidirectional segmented-memory recurrent neural network for protein secondary structure prediction
Soft Computing, 2006
The formation of protein secondary structure especially the regions of β-sheets involves long-range interactions between amino acids. We propose a novel recurrent neural network architecture called segmented-memory recurrent neural network (SMRNN) and present experimental results showing that SMRNN outperforms conventional recurrent neural networks on long-term dependency problems. In order to capture long-term dependencies in protein sequences for secondary structure prediction, we develop a predictor based on bidirectional segmented-memory recurrent neural network (BSMRNN), which is a noncausal generalization of SMRNN. In comparison with the existing predictor based on bidirectional recurrent neural network (BRNN), the BSMRNN predictor can improve prediction performance especially the recognition accuracy of β-sheets.