A dynamic Bayesian network approach to protein secondary structure prediction

Background  

Protein secondary structure prediction method based on probabilistic models such as hidden Markov model (HMM) appeals to many because it provides meaningful information relevant to sequence-structure relationship. However, at present, the prediction accuracy of pure HMM-type methods is much lower than that of machine learning-based methods such as neural networks (NN) or support vector machines (SVM).     

Prediction of the disulfide bonding state of cysteines in proteins with hidden neural networks

A hybrid system (hidden neural network) based on a hidden Markov model (HMM) and neural networks (NN) was trained to predict the bonding states of cysteines in proteins starting from the residue chains. Training was performed using 4136 cysteine-containing segments extracted from 969 non-homologous proteins of well-resolved 3D structure and without chain-breaks. After a 20-fold cross-validation procedure, the efficiency of the prediction scores as high as 80% using neural networks based on evolutionary information. When the whole protein is taken into account by means of an HMM, a hybrid system is generated, whose emission probabilities are computed using the NN output (hidden neural networks). In this case, the predictor accuracy increases up to 88%. Further, when tested on a protein basis, the hybrid system can correctly predict 84% of the chains in the data set, with a gain of at least 27% over the NN predictor.     

Prediction of transmembrane regions of beta-barrel proteins using ANN- and SVM-based methods

This article describes a method developed for predicting transmembrane beta-barrel regions in membrane proteins using machine learning techniques: artificial neural network (ANN) and support vector machine (SVM). The ANN used in this study is a feed-forward neural network with a standard back-propagation training algorithm. The accuracy of the ANN-based method improved significantly, from 70.4% to 80.5%, when evolutionary information was added to a single sequence as a multiple sequence alignment obtained from PSI-BLAST. We have also developed an SVM-based method using a primary sequence as input and achieved an accuracy of 77.4%. The SVM model was modified by adding 36 physicochemical parameters to the amino acid sequence information. Finally, ANN- and SVM-based methods were combined to utilize the full potential of both techniques. The accuracy and Matthews correlation coefficient (MCC) value of SVM, ANN, and combined method are 78.5%, 80.5%, and 81.8%, and 0.55, 0.63, and 0.64, respectively. These methods were trained and tested on a nonredundant data set of 16 proteins, and performance was evaluated using "leave one out cross-validation" (LOOCV). Based on this study, we have developed a Web server, TBBPred, for predicting transmembrane beta-barrel regions in proteins (available at http://www.imtech.res.in/raghava/tbbpred).     

HMM-based Supervised Machine Learning Framework for the Detection of fECG R-R Peak Locations

ObjectiveFetal Electro Cardiogram (fECG) provides critical information on the wellbeing of a foetus heart in its developing stages in the mother's womb. The objective of this work is to extract fECG which is buried in a composite signal consisting of itself, maternal ECG (mECG) and noises contributed from various unavoidable sources. In the past, the challenge of extracting fECG from the composite signal was dealt with by Stochastic Weiner filter, model-based Kalman filter and other adaptive filtering techniques. Blind Source Separation (BSS) based Independent Component Analysis (ICA) has shown an edge over the adaptive filtering techniques as the former does not require a reference signal. Recently, data-driven machine learning techniques e.g., adaptive neural networks, adaptive neuro-fuzzy inference system, support vector machine (SVM) are also applied.MethodThis work pursues hidden Markov model (HMM)-based supervised machine learning frame-work for the determination of the location of fECG QRS complex from the composite abdominal signal. HMM is used to model the underlying hidden states of the observable time series of the extracted and separated fECG data with its QRS peak location as one of the hidden states. The state transition probabilities are estimated in the training phase using the annotated data sets. Afterwards, using the estimated HMM networks, fQRS locations are detected in the testing phase. To evaluate the proposed technique, the accuracy of the correct detection of QRS complex with respect to the correct annotation of QRS complex location is considered and quantified by the sensitivity, probability of false alarm, and accuracy.ResultsThe best results that have been achieved using the proposed method are: accuracy – 97.1%, correct detection rate (translated to sensitivity) – 100%, and false alarm rate – 2.89%.ConclusionTwo primary challenges in these methods are finding the right reference threshold for the normalization of the extracted fECG signal during the initial trials and limitation of discrete frame work of HMM signal (converted from continuous time) which only offers a countable number of levels in observations. By feeding the posterior probabilities, obtained from SVM, into HMM, as emission probabilities, can further improve the accuracy of fQRS location detection.     

Reduced amino acid alphabet is sufficient to accurately recognize intrinsically disordered protein

Intrinsically disordered proteins are an important class of proteins with unique functions and properties. Here, we have applied a support vector machine (SVM) trained on naturally occurring disordered and ordered proteins to examine the contribution of various parameters (vectors) to recognizing proteins that contain disordered regions. We find that a SVM that incorporates only amino acid composition has a recognition accuracy of 87+/-2%. This result suggests that composition alone is sufficient to accurately recognize disorder. Interestingly, SVMs using reduced sets of amino acids based on chemical similarity preserve high recognition accuracy. A set as small as four retains an accuracy of 84+/-2%; this suggests that general physicochemical properties rather than specific amino acids are important factors contributing to protein disorder.     

BTXpred: prediction of bacterial toxins

This paper describes a method developed for predicting bacterial toxins from their amino acid sequences. All the modules, developed in this study, were trained and tested on a non-redundant dataset of 150 bacterial toxins that included 77 exotoxins and 73 endotoxins. Firstly, support vector machines (SVM) based modules were developed for predicting the bacterial toxins using amino acids and dipeptides composition and achieved an accuracy of 96.07% and 92.50%, respectively. Secondly, SVM based modules were developed for discriminating entotoxins and exotoxins, using amino acids and dipeptides composition and achieved an accuracy of 95.71% and 92.86%, respectively. In addition, modules have been developed for classifying the exotoxins (e.g. activate adenylate cyclase, activate guanylate cyclase, neurotoxins) using hidden Markov models (HMM), PSI-BLAST and a combination of the two and achieved overall accuracy of 95.75%, 97.87% and 100%, respectively. Based on the above study, a web server called 'BTXpred' has been developed, which is available at http://www.imtech.res.in/raghava/btxpred/. Supplementary information is available at http://www.imtech.res.in/raghava/btxpred/supplementary.html.     

Advances in the prediction of protein targeting signals

Enlarged sets of reference data and special machine learning approaches have improved the accuracy of the prediction of protein subcellular localization. Recent approaches report over 95% correct predictions with low fractions of false-positives for secretory proteins. A clear trend is to develop specifically tailored organism- and organelle-specific prediction tools rather than using one general method. Focus of the review is on machine learning systems, highlighting four concepts: the artificial neural feed-forward network, the self-organizing map (SOM), the Hidden-Markov-Model (HMM), and the support vector machine (SVM).     

VGIchan： Prediction and Classification of Voltage-Gated Ion Channels

This study describes methods for predicting and classifying voltage-gated ion channels. Firstly, a standard support vector machine （SVM） method was developed for predicting ion channels by using amino acid composition and dipeptide composition, with an accuracy of 82.89% and 85.56%, respectively. The accuracy of this SVM method was improved from 85.56% to 89.11% when combined with PSIBLAST similarity search. Then we developed an SVM method for classifying ion channels （potassium, sodium, calcium, and chloride） by using dipeptide composition and achieved an overall accuracy of 96.89%. We further achieved a classification accuracy of 97.78% by using a hybrid method that combines dipeptidebased SVM and hidden Markov model methods. A web server VGIchan has been developed for predicting and classifying voltage-gated ion channels using the above approaches. VGIchan is freely available at www.imtech.res.in/raghava/vgichan/.     

Protein classification based on text document classification techniques

The need for accurate, automated protein classification methods continues to increase as advances in biotechnology uncover new proteins. G-protein coupled receptors (GPCRs) are a particularly difficult superfamily of proteins to classify due to extreme diversity among its members. Previous comparisons of BLAST, k-nearest neighbor (k-NN), hidden markov model (HMM) and support vector machine (SVM) using alignment-based features have suggested that classifiers at the complexity of SVM are needed to attain high accuracy. Here, analogous to document classification, we applied Decision Tree and Naive Bayes classifiers with chi-square feature selection on counts of n-grams (i.e. short peptide sequences of length n) to this classification task. Using the GPCR dataset and evaluation protocol from the previous study, the Naive Bayes classifier attained an accuracy of 93.0 and 92.4% in level I and level II subfamily classification respectively, while SVM has a reported accuracy of 88.4 and 86.3%. This is a 39.7 and 44.5% reduction in residual error for level I and level II subfamily classification, respectively. The Decision Tree, while inferior to SVM, outperforms HMM in both level I and level II subfamily classification. For those GPCR families whose profiles are stored in the Protein FAMilies database of alignments and HMMs (PFAM), our method performs comparably to a search against those profiles. Finally, our method can be generalized to other protein families by applying it to the superfamily of nuclear receptors with 94.5, 97.8 and 93.6% accuracy in family, level I and level II subfamily classification respectively.     

Design of accurate predictors for DNA-binding sites in proteins using hybrid SVM-PSSM method

In this paper, we investigate the design of accurate predictors for DNA-binding sites in proteins from amino acid sequences. As a result, we propose a hybrid method using support vector machine (SVM) in conjunction with evolutionary information of amino acid sequences in terms of their position-specific scoring matrices (PSSMs) for prediction of DNA-binding sites. Considering the numbers of binding and non-binding residues in proteins are significantly unequal, two additional weights as well as SVM parameters are analyzed and adopted to maximize net prediction (NP, an average of sensitivity and specificity) accuracy. To evaluate the generalization ability of the proposed method SVM-PSSM, a DNA-binding dataset PDC-59 consisting of 59 protein chains with low sequence identity on each other is additionally established. The SVM-based method using the same six-fold cross-validation procedure and PSSM features has NP=80.15% for the training dataset PDNA-62 and NP=69.54% for the test dataset PDC-59, which are much better than the existing neural network-based method by increasing the NP values for training and test accuracies up to 13.45% and 16.53%, respectively. Simulation results reveal that SVM-PSSM performs well in predicting DNA-binding sites of novel proteins from amino acid sequences.     

The prediction of membrane protein structure and genome structural annotation

New methods, essentially based on hidden Markov models (HMM) and neural networks (NN), can predict the topography of both beta-barrel and all-alpha membrane proteins with high accuracy and a low rate of false positives and false negatives. These methods have been integrated in a suite of programs to filter proteomes of Gram-negative bacteria, searching for new membrane proteins.     

A probabilistic model for secondary structure prediction from protein chemical shifts

Protein chemical shifts encode detailed structural information that is difficult and computationally costly to describe at a fundamental level. Statistical and machine learning approaches have been used to infer correlations between chemical shifts and secondary structure from experimental chemical shifts. These methods range from simple statistics such as the chemical shift index to complex methods using neural networks. Notwithstanding their higher accuracy, more complex approaches tend to obscure the relationship between secondary structure and chemical shift and often involve many parameters that need to be trained. We present hidden Markov models (HMMs) with Gaussian emission probabilities to model the dependence between protein chemical shifts and secondary structure. The continuous emission probabilities are modeled as conditional probabilities for a given amino acid and secondary structure type. Using these distributions as outputs of first‐ and second‐order HMMs, we achieve a prediction accuracy of 82.3%, which is competitive with existing methods for predicting secondary structure from protein chemical shifts. Incorporation of sequence‐based secondary structure prediction into our HMM improves the prediction accuracy to 84.0%. Our findings suggest that an HMM with correlated Gaussian distributions conditioned on the secondary structure provides an adequate generative model of chemical shifts. Proteins 2013; © 2012 Wiley Periodicals, Inc.     

An SVM method using evolutionary information for the identification of allergenic proteins

This study presents an allergenic protein prediction system that appears to be capable of producing high sensitivity and specificity. The proposed system is based on support vector machine (SVM) using evolutionary information in the form of an amino acid position specific scoring matrix (PSSM). The performance of this system is assessed by a 10-fold cross-validation experiment using a dataset consisting of 693 allergens and 1041 non-allergens obtained from Swiss-Prot and Structural Database of Allergenic Proteins (SDAP). The PSSM method produced an accuracy of 90.1% in comparison to the methods based on SVM using amino acid, dipeptide composition, pseudo (5-tier) amino acid composition that achieved an accuracy of 86.3, 86.5 and 82.1% respectively. The results show that evolutionary information can be useful to build more effective and efficient allergen prediction systems.     

基于间隔二肽组分和递归特征消除法的DNA结合蛋白的鉴定

DNA结合蛋白(DNA-binding proteins,DBPs)的鉴定在原核和真核生物的基因和蛋白质功能注释研究中具有十分重要的意义.本研究首次运用间隔二肽组分(gapped-dipeptide composition,Gap DPC)结合递归特征消除法(recursive feature elimination,RFE)鉴定DBPs.首先获得待测蛋白质氨基酸序列的位置特异性得分矩阵(position specific scoring matrix,PSSM),在此基础上提取蛋白质的Gap DPC特征,通过RFE法选择最优特征,然后利用支持向量机(support vector machine,SVM)作为分类器,在蛋白质序列数据集PDB396和LB1068中进行夹克刀交叉验证(jackknife cross validation test).研究结果显示,基于PDB396和LB1068数据集,DBPs预测的准确率、Matthews相关系数、敏感性和特异性分别达到93.43%、0.86、89.04%和96.00%,以及86.33%、0.73、86.49%和86.18%,明显优于文献报道中的相关方法,为DBPs的鉴定提供了新的模型.     

GNBSL: a new integrative system to predict the subcellular location for Gram-negative bacteria proteins

This paper proposes a new integrative system (GNBSL--Gram-negative bacteria subcellular localization) for subcellular localization specifized on the Gram-negative bacteria proteins. First, the system generates a position-specific frequency matrix (PSFM) and a position-specific scoring matrix (PSSM) for each protein sequence by searching the Swiss-Prot database. Then different features are extracted by four modules from the PSFM and the PSSM. The features include whole-sequence amino acid composition, N- and C-terminus amino acid composition, dipeptide composition, and segment composition. Four probabilistic neural network (PNN) classifiers are used to classify these modules. To further improve the performance, two modules trained by support vector machine (SVM) are added in this system. One module extracts the residue-couple distribution from the amino acid sequence and the other module applies a pairwise profile alignment kernel to measure the local similarity between every two sequences. Finally, an additional SVM is used to fuse the outputs from the six modules. Test on a benchmark dataset shows that the overall success rate of GNBSL is higher than those of PSORT-B, CELLO, and PSLpred. A web server GNBSL can be visited from http://166.111.24.5/webtools/GNBSL/index.htm.     

A machine learning based method for the prediction of secretory proteins using amino acid composition, their order and similarity-search

Most of the prediction methods for secretory proteins require the presence of a correct N-terminal end of the preprotein for correct classification. As large scale genome sequencing projects sometimes assign the 5'-end of genes incorrectly, many proteins are encoded without the correct N-terminus leading to incorrect prediction. In this study, a systematic attempt has been made to predict secretory proteins irrespective of presence or absence of N-terminal signal peptides (also known as classical and non-classical secreted proteins respectively), using machine-learning techniques; artificial neural network (ANN) and support vector machine (SVM). We trained and tested our methods on a dataset of 3321 secretory and 3654 non-secretory mammalian proteins using five-fold cross-validation technique. First, ANN-based modules have been developed for predicting secretory proteins using 33 physico-chemical properties, amino acid composition and dipeptide composition and achieved accuracies of 73.1%, 76.1% and 77.1%, respectively. Similarly, SVM-based modules using 33 physico-chemical properties, amino acid, and dipeptide composition have been able to achieve accuracies of 77.4%, 79.4% and 79.9%, respectively. In addition, BLAST and PSI-BLAST modules designed for predicting secretory proteins based on similarity search achieved 23.4% and 26.9% accuracy, respectively. Finally, we developed a hybrid-approach by integrating amino acid and dipeptide composition based SVM modules and PSI-BLAST module that increased the accuracy to 83.2%, which is significantly better than individual modules. We also achieved high sensitivity of 60.4% with low value of 5% false positive predictions using hybrid module. A web server SRTpred has been developed based on above study for predicting classical and non-classical secreted proteins from whole sequence of mammalian proteins, which is available from http://www.imtech.res.in/raghava/srtpred/.     

Recognition and classification of histones using support vector machine.

Histones are DNA-binding proteins found in the chromatin of all eukaryotic cells. They are highly conserved and can be grouped into five major classes: H1/H5, H2A, H2B, H3, and H4. Two copies of H2A, H2B, H3, and H4 bind to about 160 base pairs of DNA forming the core of the nucleosome (the repeating structure of chromatin) and H1/H5 bind to its DNA linker sequence. Overall, histones have a high arginine/lysine content that is optimal for interaction with DNA. This sequence bias can make the classification of histones difficult using standard sequence similarity approaches. Therefore, in this paper, we applied support vector machine (SVM) to recognize and classify histones on the basis of their amino acid and dipeptide composition. On evaluation through a five-fold cross-validation, the SVM-based method was able to distinguish histones from nonhistones (nuclear proteins) with an accuracy around 98%. Similarly, we obtained an overall >95% accuracy in discriminating the five classes of histones through the application of 1-versus-rest (1-v-r) SVM. Finally, we have applied this SVM-based method to the detection of histones from whole proteomes and found a comparable sensitivity to that accomplished by hidden Markov motifs (HMM) profiles.