RS126 Data Set Research Articles

Protein Secondary Structure Prediction based on CNN and Machine Learning Algorithms

One of the most important topics in computational biology is protein secondary structure prediction. Primary, secondary, tertiary, and quaternary structure are the four levels of complexity that can be used to characterize the entire structure of a protein that are totally ordered by the amino acid sequences. The polypeptide backbone of a protein's local configuration is referred to as a secondary structure. In this paper, three prediction algorithms have been proposed which will predict the protein secondary structure based on machine learning. These prediction methods have been improved by the model structure of convolutional neural networks (CNN). The Rectified Linear Units (ReLU) has been used as the activation function. The 2D CNN has been trained with machine learning algorithms, including Support Vector Machine, Naive Bays and Random Forest. The SVM is used to correctly classify the unseen data. Naïve Bays (NB) and Random Forest (RF) are also applied to solve the prediction problems for not only classification problems but also regression problems. The 2D CNN, hybrid of 2D CNN -SVM, CNN-RF and CNN-NB have been proposed in this experiment. These different methods are implemented with the RS126, 25PDB and CB513 dataset. Further, all prediction Q3 accuracy is compared and improved with their datasets.

Protein secondary structure prediction (PSSP) using different machine algorithms

BackgroundThe computational biology approach has advanced exponentially in protein secondary structure prediction (PSSP), which is vital for the pharmaceutical industry. Extracting protein structure from the laboratory has insufficient information for PSSP that is used in bioinformatics studies. In this paper, the support vector machine (SVM) model and decision tree are presented on the RS126 dataset to address the problem of PSSP. A decision tree is applied for the SVM outcome to obtain the relevant guidelines possible for PSSP. Furthermore, the number of produced rules was fairly small, and they show a greater degree of comprehensibility compared to other rules. Several of the proposed principles have compelling and relevant biological clarification.ResultsThe results confirmed that the existence of a particular amino acid in a protein sequence increases the stability for the forecast of protein secondary structure. The suggested algorithm achieved 85% accuracy for the E|~E classifier.ConclusionsThe proposed rules can be very important in managing wet laboratory experiments intended at determining protein secondary structure. Lastly, future work will focus mainly on large protein datasets without overfitting and expand the amount of extracted regulations for PSSP.

A Meta-heuristic Framework for Secondary Protein Structure Prediction using BAT-FLANN Optimization Algorithm

Background/Objectives: Proteins are the fundamental units of biology; the mechanism by which primary sequence of proteins is predicted into its secondary structure is not yet accurately achieved. Methods/Statistical analysis: In this paper, BAT inspired FLANN (Functional Link Artificial Neural Network) model for protein secondary structure prediction with low computation cost and accuracy has been proposed. The proposed model consists of three different phases; i) First, the primary sequence of amino acid is converted into dynamic matrix for different window sizes then this dynamic matrix is used to derive correlation matrix, ii) Second, FLANN is used to classify each sequence of correlation matrix with different learning parameters and random weights. BAT inspired optimization algorithm has been used to optimize the weight and learning parameters of BAT-FLANN, and (iii) finally, refinement of secondary structure result. Results: Experiments were conducted with real datasets of some primary sequence on RS126 and CB396 datasets. Proposed method has been compared with existing DSC, NNSSP, PHD, PREDATOR, ZPRED, MULPRED, SVM models and found to be more promising. Conclusion/Application: The proposed method achieves average Q3 accuracy 81.2% and 82.7% for CB396 and RS126 dataset respectively. Moreover the segment overlap (SOV) is 76.1% and 75.3% for CB396 and RS126 dataset respectively.

Effect of simple ensemble methods on protein secondary structure prediction

Ensemble methods for building improved classifier models have been an important topic in machine learning, pattern recognition and data mining areas, where they have shown great promise. They boast a robustness that has spearheaded their application in many practical classification problems, especially when there is a significant diversity among the ensemble members. Actually, they replace traditional machine learning techniques in many applications and special attention has been devoted to them as a mean to improve the prediction accuracy for problems of high complexity. Several combination rules have been investigated in this context. However, it is claimed that no rule is always better than others for designing an optimal decision. The present study evaluates the performance of two different ensemble methods for protein secondary structure prediction. We focus on weighted opinions pooling and the most common aggregation rules for decisions inference. The ensemble members are accurate protein secondary structure single model predictors namely, Multi-Class Support Vector Machines and Artificial Neural Networks. Experiments are carried out using cross-validation tests on RS126 and CB513 benchmark datasets. Our results clearly confirm that ensembles are more accurate than a single model and the experimental comparison of the investigated ensemble schemes demonstrates that the newly introduced rule called Exponential Opinion Pool competes well against state-of-the-art fixed rules, especially the sum rule which in some cases is able to achieve better performance.

The Hybrid Method of Fuzzy Feed-Forward Neural Network for Predicting Protein Secondary Structure

With respect to the fact that the prediction of Protein secondary structure based on amino acids is very important, therefore, this study tries to present a new method based on the fuzzy combinational structure of a set of feed-forward neural networks so that the prediction accuracy of Protein secondary structure can be improved compared with the existing methods. Neural networks used in this paper are based on time windows; also, different methods have been established and trained to infer the three states of alpha- helix, beta- sheet and coils from DSSP results, and finally, combining the results of the abovementioned networks in a fuzzy manner, the prediction method of Protein secondary structure based on neural network has been improved. It should be noted that in this paper, CB513 and RS126 data sets which are valid data sets in evaluating prediction methods of Protein secondary structure known in research studies in this area have been used to train and evaluate the proposed method.

Predicting the Secondary Structure of Proteins by Cascading Neural Networks

Protein Secondary Structure Prediction (PSSP) is considered as a challenging task in bioinformatics and so many approaches have been prop osed in the literature to solve this problem via achieving more accurate prediction results. Accurate prediction of secondary structure is a critical role in deducing tertiary structure of proteins and their functions. Among the proposed approaches to tack le this problem, Artificial Neural Networks (ANNs) are considered as one of the successful tools that are widely used in this field. Recently, many efforts have been devoted to modify, improve and combine this methodology with other machine learning method s in order to get better results. In this work, we have proposed a two -stage feed forward neural network for prediction of protein secondary structures. To evaluate our approach, it is applied on RS126 dataset and its results are compared with some other N-based methods.

Ensemble of Neural Networks to Solve Class Imbalance Problem of Protein Secondary Structure Prediction

Protein secondary structures prediction (PSSP) is considered as a challenging task in bioinformatics. Many approaches have been proposed in last few decades in order to solve this problem. Despite the enhancements achieved, the prediction accuracy still remains limited. Accurate prediction of the secondary structure of proteins is a critical step in deducing tertiary structure of proteins and their functions. Among the proposed approaches to tackle this problem, artificial neural networks (ANN) are considered as one of the most successful methods that widely used in the field of PSSP. Recently, many efforts have been devoted to modify, improve and combine this technology with other machine learning methods in order to get better results. In this paper, we have proposed an ensemble method which combines the outputs of four feedforward neural networks. In each network one of the machine learning approaches has been applied in order to solve the class imbalance problem of protein secondary structure classification. The experimental results on RS126 data set show that our ensemble system has better performance compare to the best individual classifier. The results also reveal that the proposed system yields significant improvement in prediction accuracy of beta-sheet structure and a more balanced classification of three secondary structures.

Protein secondary structure prediction using DWKF based on SVR-NSGAII

Prediction of protein secondary structure is an important step towards elucidating its three dimensional structure and its function. This is a challenging problem in bioinformatics. By introduction of machine learning for protein structure prediction, a solution has brought to this challenge to some extent. In the literature of Machine learning or data mining, regression and classification problems are typically viewed as two distinct problems differentiated by continuous or categorical dependent variable. There are endeavors to use regression methods to solve the classification problem and vice versa. To regard a classification problem as a regression one, we proposed a method which is based on Support Vector Regression (SVR) classification model as one of the powerful methods in the field of machine intelligence. We applied non-dominated Sorting Genetic Algorithm II (NSGAII) to find mapping points (MPs) for rounding a real-value to an integer one. Also NSGAII is used for finding out and tuning SVR kernel parameters optimally to enhance the performance of our model and achieve better results. At the other hand, using a suitable SVR kernel function for a particular problem can improve the prediction results remarkably but there is not a kernel which can predict all protein secondary structure classes with acceptable accuracy. Therefore we use a Dynamic Weighted Kernel Fusion (DWKF) method for fusing of three SVR kernels to achieve a supreme performance. Also to improve our method, Position Scoring Matrix (PSSM) profiles are used as the input information to it. The goals of this research are to regulate SVR parameters and fuse different SVR kernel outputs in order to determine protein secondary structure classes accurately. The obtained classification accuracies of our method are 85.79% and 84.94% on RS126 and CB513 datasets respectively and they are promising with regard to other classification methods in the literature. Moreover, for gauging our method behavior in comparison to other state of arts methods, an independent dataset is used and achieves 81.4% accuracy. Our method cannot achieve the best value for any considered performance metrics on an independent dataset but its values for whole metrics are quite acceptable.

A new intelligent prediction system model-the compound pyramid model

A current development trend in research on intelligent systems is to optimize a general intelligent prediction system into an individuation intelligent prediction system that is applied in specialized fields. Protein structure prediction is a challenging international issue. In this paper, we propose a new intelligent prediction system model, designed as a multi-layer compound pyramid model, for predicting secondary protein structure. The model comprises four independent intelligent interfaces and several knowledge discovery methods. The model penetrates throughout the domain knowledge, with the effective attributes chosen by Causal Cellular Automata. Furthermore, a high pure structure database is constructed for training. On the RS126 dataset, the overall state per-residue accuracy, Q3, reached 83.99%, while on the CB513 dataset, Q3 reached 85.58%. Meanwhile, on the CASP8 sequences, the results are superior to those produced by other methods, such as Psipred, Jpred, APSSP2 and BehairPred. These results confirm that our method has a strong generalization ability, and that it provides a model for the construction of other intelligent systems.

PSSP with dynamic weighted kernel fusion based on SVM-PHGS

Since 1960s, researchers have proposed several prediction methods, for protein secondary structure prediction (PSSP), whereas the accuracy of them is no more than 80%. In this case, there is an urgent need to introduce a high accuracy prediction method. One learning method called support vector machines (SVMs) has shown comparable or better results than neural networks on bioinformatics applications. This research proposes a method based on SVM which has been improved by a new parallel multi class (PMC) method, parallel hierarchical grid search (PHGS), cross validation (CV) technique and weighted kernel fusion (WKF) method. The presented PHGS has been applied to regularize parameters of SVM’s kernel function which have an important impact on the accuracy. Using a suitable input data and kernel function for a particular problem can improve the prediction results remarkably. Also to improve our method, Position Scoring Matrix (PSSM) profiles are used as the input information to it. The goals of this study are to calibrate kernel function parameters and fusion of different kernel functions’ result in order to determine protein secondary structure classes accurately. The right choice of a fusion method is an important issue in creating a supreme performance so we propose a dynamic weight allocation method based on a non-linear analysis system. The obtained classification accuracies of our method are 84.65% and 83.94% on RS126 and CB513 datasets respectively and they are very promising with regard to other classification methods in the literature for this problem. Also for evaluating our method behavior in comparison to other state of arts methods, an independent dataset is used. The results show that the comprehensibility of WKF based on SVM-PHGS is much better than other methods.

HYBP_PSSP: a hybrid back propagation method for predicting protein secondary structure

The prediction of secondary structure is an important topic in the field of bioinformatics, even if the methods have matured, and development of the algorithms is a far less active area than a decade ago. Accurate prediction is very useful to biologists in its own right, but it is worth pointing out that it is also an essential component of tertiary structure prediction, which in contrast is far from solved and continues to be a highly active area of research. In addition, sequence comparison methods have more recently incorporated local structure tracks. The extra information utilized by the new methods has led to considerable improvements in fold recognition and alignment accuracy. In this paper, a novel method for protein secondary structure prediction is presented. Using evolutionary information contained in amino acid’s physicochemical properties, position-specific scoring matrix generated by PSI-BLAST and HMMER3 profiles as input to hybrid back propagation system, secondary structure can be predicted at significantly increased accuracy. Based on knowledge discovery theory based on inner cognitive mechanism (KDTICM) theory, we have constructed a compound pyramid model approach, which is composed of four layers of the intelligent interface and integrated in several ways, such as hybrid back propagation method (HBP), modified knowledge discovery in databases (KDD*), hybrid SVM method (HSVM) and so on. Experiments on three standard datasets (RS126, CB513 and CASP8) show that CPM is capable of producing the higher Q 3 and SOV scores than that achieved by existing widely used schemes such as PSIPRED, PHD, Predator, as well as previously developed prediction methods. On the RS126 and CB513 datasets, it achieves a Q 3 and SOV99 score are considerably higher than the best reported scores, respectively. It is also tested on target proteins of critical assessment of protein structure prediction experiment (CASP8) and achieves better results than the traditional methods, including the popular PSIPRED method over overall prediction accuracy. Available: http://www.kdd.ustb.edu.cn/protein_Web.

Protein 8‐class secondary structure prediction using conditional neural fields

Compared with the protein 3-class secondary structure (SS) prediction, the 8-class prediction gains less attention and is also much more challenging, especially for proteins with few sequence homologs. This paper presents a new probabilistic method for 8-class SS prediction using conditional neural fields (CNFs), a recently invented probabilistic graphical model. This CNF method not only models the complex relationship between sequence features and SS, but also exploits the interdependency among SS types of adjacent residues. In addition to sequence profiles, our method also makes use of non-evolutionary information for SS prediction. Tested on the CB513 and RS126 data sets, our method achieves Q8 accuracy of 64.9 and 64.7%, respectively, which are much better than the SSpro8 web server (51.0 and 48.0%, respectively). Our method can also be used to predict other structure properties (e.g. solvent accessibility) of a protein or the SS of RNA.

Improving protein secondary structure prediction using a multi-modal BP method

Methods for predicting protein secondary structures provide information that is useful both in ab initio structure prediction and as additional restraints for fold recognition algorithms. Secondary structure predictions may also be used to guide the design of site directed mutagenesis studies, and to locate potential functionally important residues. In this article, we propose a multi-modal back propagation neural network (MMBP) method for predicting protein secondary structures. Using a Knowledge Discovery Theory based on Inner Cognitive Mechanism (KDTICM) method, we have constructed a compound pyramid model (CPM), which is composed of three layers of intelligent interface that integrate multi-modal back propagation neural network (MMBP), mixed-modal SVM (MMS), modified Knowledge Discovery in Databases (KDD ⁎) process and so on. The CPM method is both an integrated web server and a standalone application that exploits recent advancements in knowledge discovery and machine learning to perform very accurate protein secondary structure predictions. Using a non-redundant test dataset of 256 proteins from RCASP256, the CPM method achieves an average Q 3 score of 86.13% (SOV99=84.66%). Extensive testing indicates that this is significantly better than any other method currently available. Assessments using RS126 and CB513 datasets indicate that the CPM method can achieve average Q 3 score approaching 83.99% (SOV99=80.25%) and 85.58% (SOV99=81.15%). By using both sequence and structure databases and by exploiting the latest techniques in machine learning it is possible to routinely predict protein secondary structure with an accuracy well above 80%. A program and web server, called CPM, which performs these secondary structure predictions, is accessible at http://kdd.ustb.edu.cn/protein_Web/.

PSP_MCSVM: brainstorming consensus prediction of protein secondary structures using two-stage multiclass support vector machines

Secondary structure prediction is a crucial task for understanding the variety of protein structures and performed biological functions. Prediction of secondary structures for new proteins using their amino acid sequences is of fundamental importance in bioinformatics. We propose a novel technique to predict protein secondary structures based on position-specific scoring matrices (PSSMs) and physico-chemical properties of amino acids. It is a two stage approach involving multiclass support vector machines (SVMs) as classifiers for three different structural conformations, viz., helix, sheet and coil. In the first stage, PSSMs obtained from PSI-BLAST and five specially selected physicochemical properties of amino acids are fed into SVMs as features for sequence-to-structure prediction. Confidence values for forming helix, sheet and coil that are obtained from the first stage SVM are then used in the second stage SVM for performing structure-to-structure prediction. The two-stage cascaded classifiers (PSP_MCSVM) are trained with proteins from RS126 dataset. The classifiers are finally tested on target proteins of critical assessment of protein structure prediction experiment-9 (CASP9). PSP_MCSVM with brainstorming consensus procedure performs better than the prediction servers like Predator, DSC, SIMPA96, for randomly selected proteins from CASP9 targets. The overall performance is found to be comparable with the current state-of-the art. PSP_MCSVM source code, train-test datasets and supplementary files are available freely in public domain at: http://sysbio.icm.edu.pl/secstruct and http://code.google.com/p/cmater-bioinfo/ Electronic supplementary materialThe online version of this article (doi:10.1007/s00894-011-1102-8) contains supplementary material, which is available to authorized users.

Toward Better Understanding of Protein Secondary Structure: Extracting Prediction Rules

Although numerous computational techniques have been applied to predict protein secondary structure (PSS), only limited studies have dealt with discovery of logic rules underlying the prediction itself. Such rules offer interesting links between the prediction model and the underlying biology. In addition, they enhance interpretability of PSS prediction by providing a degree of transparency to the predicting model usually regarded as a black box. In this paper, we explore the generation and use of C4.5 decision trees to extract relevant rules from PSS predictions modeled with two-stage support vector machines (TS-SVM). The proposed rules were derived on the RS126 data set of 126 nonhomologous globular proteins and on the PSIPRED data set of 1,923 protein sequences. Our approach has produced sets of comprehensible, and often interpretable, rules underlying the PSS predictions. Moreover, many of the rules seem to be strongly supported by biological evidence. Further, our approach resulted in good prediction accuracy, few and usually compact rules, and rules that are generally of higher confidence levels than those generated by other rule extraction techniques.

Predicting protein second structure using a novel hybrid method

Accurate protein secondary structure predictions play an important role for direct tertiary structure modeling, and it also can significantly improve sequence analysis and sequence-structure threading for aiding in structure and function determination. Hence the improvement of predictive accuracy of the secondary structure prediction becomes essential for future development of the whole field of protein research. In this article, we propose a gradually enhanced, multi-layered prediction systematic model to predict protein secondary structure, Compound Pyramid Model (CPM). This model is composed of four independent coordination’s layers by intelligent interfaces, synthesizes several methods, such as KDD ∗, mixed-modal SVM method, mixed-modal BP method and so on. The model penetrates the whole domain knowledge, and the effective physicochemical properties of amino acids are imported. On the RS126 data set, state overall per-residue accuracy, Q 3, reached 83.99%, while segment overlap (SOV99) accuracy increased to 80.6%. On the CB513 data set, Q 3 reached 85.58%, SOV99 accuracy increased to 79.84%. Meanwhile, the results are found to be superior to those produced by other methods with blind test dataset CASP8’s sequences, including the popular Psipred method according to Q 3 and SOV99 accuracy. The result shows that our method has strong generalization ability.

Prediction of Protein Secondary Structure with two-stage multi-class SVMs

Bioinformatics techniques to Protein Secondary Structure (PSS) prediction mostly depend on the information available in amino acid sequences. In this paper, we propose a two-stage Multi-class Support Vector Machine (MSVM) approach, where the second MSVM predictor is introduced at the output of the first stage MSVM to capture the contextual relationship among secondary structure elements in order to minimise the generalisation error in the prediction. By using position-specific scoring matrices generated by PSI-BLAST, the two-stage MSVM approach achieves Q3 accuracies of 78.0% and 76.3% on the RS126 dataset of 126 non-homologous globular proteins and the CB396 dataset of 396 non-homologous proteins, respectively, which are better than the scores reported on both datasets to date. By using MSVM, the present prediction scheme significantly achieves 2-6% and 3-15% of improvement in Q3 and Sov accuracies, respectively, on the two datasets. On larger blind-test datasets from PSIPRED, CASP4 and EVA datasets, two-stage MSVM approach achieves Q3 accuracies from 77.0% to 79.5%.

PROTEIN SECONDARY STRUCTURE PREDICTION USING SUPPORT VECTOR MACHINES AND A NEW FEATURE REPRESENTATION

Knowledge of the secondary structure and solvent accessibility of a protein plays a vital role in the prediction of fold, and eventually the tertiary structure of the protein. A challenging issue of predicting protein secondary structure from sequence alone is addressed. Support vector machines (SVM) are employed for the classification and the SVM outputs are converted to posterior probabilities for multi-class classification. The effect of using Chou–Fasman parameters and physico-chemical parameters along with evolutionary information in the form of position specific scoring matrix (PSSM) is analyzed. These proposed methods are tested on the RS126 and CB513 datasets. A new dataset is curated (PSS504) using recent release of CATH. On the CB513 dataset, sevenfold cross-validation accuracy of 77.9% was obtained using the proposed encoding method. A new method of calculating the reliability index based on the number of votes and the Support Vector Machine decision value is also proposed. A blind test on the EVA dataset gives an average Q3accuracy of 74.5% and ranks in top five protein structure prediction methods. Supplementary material including datasets are available on .

Rule Generation for Protein Secondary Structure Prediction With Support Vector Machines and Decision Tree

Support vector machines (SVMs) have shown strong generalization ability in a number of application areas, including protein structure prediction. However, the poor comprehensibility hinders the success of the SVM for protein structure prediction. The explanation of how a decision made is important for accepting the machine learning technology, especially for applications such as bioinformatics. The reasonable interpretation is not only useful to guide the "wet experiments," but also the extracted rules are helpful to integrate computational intelligence with symbolic AI systems for advanced deduction. On the other hand, a decision tree has good comprehensibility. In this paper, a novel approach to rule generation for protein secondary structure prediction by integrating merits of both the SVM and decision tree is presented. This approach combines the SVM with decision tree into a new algorithm called SVM_ DT, which proceeds in three steps. This algorithm first trains an SVM. Then, a new training set is generated through careful selection from the output of the SVM. Finally, the obtained training set is used to train a decision tree learning system and to extract the corresponding rule sets. The results of the experiments of protein secondary structure prediction on RS126 data set show that the comprehensibility of SVM_DT is much better than that of the SVM. Moreover, the generalization ability of SVM_DT is better than that of C4.5 decision trees and is similar to that of the SVM. Hence, SVM_DT can be used not only for prediction, but also for guiding biological experiments.

Prediction of protein solvent accessibility using fuzzy k-nearest neighbor method

The solvent accessibility of amino acid residues plays an important role in tertiary structure prediction, especially in the absence of significant sequence similarity of a query protein to those with known structures. The prediction of solvent accessibility is less accurate than secondary structure prediction in spite of improvements in recent researches. The k-nearest neighbor method, a simple but powerful classification algorithm, has never been applied to the prediction of solvent accessibility, although it has been used frequently for the classification of biological and medical data. We applied the fuzzy k-nearest neighbor method to the solvent accessibility prediction, using PSI-BLAST profiles as feature vectors, and achieved high prediction accuracies. With leave-one-out cross-validation on the ASTRAL SCOP reference dataset constructed by sequence clustering, our method achieved 64.1% accuracy for a 3-state (buried/intermediate/exposed) prediction (thresholds of 9% for buried/intermediate and 36% for intermediate/exposed) and 86.7, 82.0, 79.0 and 78.5% accuracies for 2-state (buried/exposed) predictions (thresholds of each 0, 5, 16 and 25% for buried/exposed), respectively. Our method also showed slightly better accuracies than other methods by about 2-5% on the RS126 dataset and a benchmarking dataset with 229 proteins. Program and datasets are available at http://biocom1.ssu.ac.kr/FKNNacc/ jul@ssu.ac.kr.