A critical assessment of the secondary structure alpha-helices and their termini in proteins

Secondary structure prediction from amino acid sequence is a key component of protein structure prediction, with current accuracy at approximately 75%. We analysed two state-of-the-art secondary structure prediction methods, PHD and JPRED, comparing predictions with secondary structure assigned by the algorithms DSSP and STRIDE. The specific focus of our study was alpha-helix N-termini, as empirical free energy scales are available for residue preferences at N-terminal positions. Although these prediction methods perform well in general at predicting the alpha-helical locations and length distributions in proteins, they perform less well at predicting the correct helical termini. For example, although most predicted alpha-helices overlap a real alpha-helix (with relatively few completely missed or extra predicted helices), only one-third of JPRED and PHD predictions correctly identify the N-terminus. Analysis of neighbouring N-terminal sequences to predicted helical N-termini shows that the correct N-terminus is often within one or two residues. More importantly, the true N-terminal motif is, on average, more favourable as judged by our experimentally measured free energies. This suggests a simple, but powerful, strategy to improve secondary structure prediction using empirically derived energies to adjust the predicted output to a more favourable N-terminal sequence.     

PROFcon: novel prediction of long-range contacts

MOTIVATION: Despite the continuing advance in the experimental determination of protein structures, the gap between the number of known protein sequences and structures continues to increase. Prediction methods can bridge this sequence-structure gap only partially. Better predictions of non-local contacts between residues could improve comparative modeling, fold recognition and could assist in the experimental structure determination. RESULTS: Here, we introduced PROFcon, a novel contact prediction method that combines information from alignments, from predictions of secondary structure and solvent accessibility, from the region between two residues and from the average properties of the entire protein. In contrast to some other methods, PROFcon predicted short and long proteins at similar levels of accuracy. As expected, PROFcon was clearly less accurate when tested on sparse evolutionary profiles, that is, on families with few homologs. Prediction accuracy was highest for proteins belonging to the SCOP alpha/beta class. PROFcon compared favorably with state-of-the-art prediction methods at the CASP6 meeting. While the performance may still be perceived as low, our method clearly pushed the mark higher. Furthermore, predictions are already accurate enough to seed predictions of global features of protein structure.     

Combining evolutionary information and neural networks to predict protein secondary structure

Using evolutionary information contained in multiple sequence alignments as input to neural networks, secondary structure can be predicted at significantly increased accuracy. Here, we extend our previous three-level system of neural networks by using additional input information derived from multiple alignments. Using a position-specific conservation weight as part of the input increases performance. Using the number of insertions and deletions reduces the tendency for overprediction and increases overall accuracy. Addition of the global amino acid content yields a further improvement, mainly in predicting structural class. The final network system has a sustained overall accuracy of 71.6% in a multiple cross-validation test on 126 unique protein chains. A test on a new set of 124 recently solved protein structures that have no significant sequence similarity to the learning set confirms the high level of accuracy. The average cross-validated accuracy for all 250 sequence-unique chains is above 72%. Using various data sets, the method is compared to alternative prediction methods, some of which also use multiple alignments: the performance advantage of the network system is at least 6 percentage points in three-state accuracy. In addition, the network estimates secondary structure content from multiple sequence alignments about as well as circular dichroism spectroscopy on a single protein and classifies 75% of the 250 proteins correctly into one of four protein structural classes. Of particular practical importance is the definition of a position-specific reliability index. For 40% of all residues the method has a sustained three-state accuracy of 88%, as high as the overall average for homology modelling. A further strength of the method is greatly increased accuracy in predicting the placement of secondary structure segments. © 1994 Wiley-Liss, Inc.     

Prediction of the location and type of beta-turns in proteins using neural networks.

A neural network has been used to predict both the location and the type of beta-turns in a set of 300 nonhomologous protein domains. A substantial improvement in prediction accuracy compared with previous methods has been achieved by incorporating secondary structure information in the input data. The total percentage of residues correctly classified as beta-turn or not-beta-turn is around 75% with predicted secondary structure information. More significantly, the method gives a Matthews correlation coefficient (MCC) of around 0.35, compared with a typical MCC of around 0.20 using other beta-turn prediction methods. Our method also distinguishes the two most numerous and well-defined types of beta-turn, types I and II, with a significant level of accuracy (MCCs 0.22 and 0.26, respectively).     

Improved performance in protein secondary structure prediction by combining multiple predictions

In this paper(1) we present a novel framework for protein secondary structure prediction. In this prediction framework, firstly we propose a novel parameterized semi-probability profile, which combines single sequence with evolutionary information effectively. Secondly, different semi-probability profiles are respectively applied as network input to predict protein secondary structure. Then a comparison among these different predictions is discussed in this article. Finally, na?ve Bayes approaches are used to combine these predictions in order to obtain a better prediction performance than individual prediction. The experimental results show that our proposed framework can indeed improve the prediction accuracy.     

Protein structure prediction

Current methods developed for predicting protein structure are reviewed. The most widely used algorithms of Chou and Fasman and Garnier et al for predicting secondary structure are compared to the most recent ones including sequence similarity methods, neural network, pattern recognition or joint prediction methods. The best of these methods correctly predict 63-65% of the residues in the database with cross-validation for 3 conformations, helix, beta strand and coli with a standard deviation of 6-8% per protein. However, when a homologous protein is already in the database, the accuracy of prediction by the similarity peptide method of Levin and Garnier reaches about 90%. Some conclusions can be drawn on the mechanism of protein folding. As all the prediction methods only use the local sequence for prediction (+/- 8 residues maximum) one can infer that 65% of the conformation of a residue is dictated on average by the local sequence, the rest is brought by the folding. The best predicted proteins or peptide segments are those for which the folding has less effect on the conformation. Presently, prediction of tertiary structure is only of practical use when the structure of a homologous protein is already known. Amino acid alignment to define residues of equivalent spatial position is critical for modelling of the protein. We showed for serine proteases that secondary structure prediction can help to define a better alignment. Non-homologous segments of the polypeptide chain, such as loops, libraries of known loops and/or energy minimization with various force fields, are used without yet giving satisfactory solutions. An example of modelling by homology, aided by secondary structure prediction on 2 regulatory proteins, Fnr and FixK is presented.     

Secondary structural analysis of retrovirus integrase: characterization by circular dichroism and empirical prediction methods

The retrovirus integrase (IN) protein is essential for integration of viral DNA into host DNA. The secondary structure of the purified IN protein from avian myeloblastosis virus was investigated by both circular dichroism (CD) spectroscopy and five empirical prediction methods. The secondary structures determined from the resolving of CD spectra through a least-squares curve fitting procedure were compared with those predicted from four statistical methods, e.g., the Chou-Fasman, Garnier-Osguthorpe-Robson, Nishikawa-Ooi, and a JOINT scheme which combined all three of these methods, plus a pure a priori one, the Ptitsyn-Finkelstein method. Among all of the methods used, the Nishikawa-Ooi prediction gave the closest match in the composition of secondary structure to the CD result, although the other methods each correctly predicted one or more secondary structural group. Most of the alpha-helix and beta-sheet states predicted by the Ptitsyn-Finkelstein method were in accord with the Nishikawa-Ooi method. Secondary structural predictions by the Nishikawa-Ooi method were extended further to include IN proteins from four phylogenetic distinct retroviruses. The structural relationships between the four most conserved amino acid blocks of these IN proteins were compared using sequence homology and secondary structure predictions.     

An algorithm for protein secondary structure prediction based on class prediction

An algorithm has been developed to improve the success rate in the prediction of the secondary structure of proteins by taking into account the predicted class of the proteins. This method has been called the 'double prediction method' and consists of a first prediction of the secondary structure from a new algorithm which uses parameters of the type described by Chou and Fasman, and the prediction of the class of the proteins from their amino acid composition. These two independent predictions allow one to optimize the parameters calculated over the secondary structure database to provide the final prediction of secondary structure. This method has been tested on 59 proteins in the database (i.e. 10,322 residues) and yields 72% success in class prediction, 61.3% of residues correctly predicted for three states (helix, sheet and coil) and a good agreement between observed and predicted contents in secondary structure.     

Accurate prediction of protein structural classes using functional domains and predicted secondary structure sequences

Protein structural class prediction is one of the challenging problems in bioinformatics. Previous methods directly based on the similarity of amino acid (AA) sequences have been shown to be insufficient for low-similarity protein data-sets. To improve the prediction accuracy for such low-similarity proteins, different methods have been recently proposed that explore the novel feature sets based on predicted secondary structure propensities. In this paper, we focus on protein structural class prediction using combinations of the novel features including secondary structure propensities as well as functional domain (FD) features extracted from the InterPro signature database. Our comprehensive experimental results based on several benchmark data-sets have shown that the integration of new FD features substantially improves the accuracy of structural class prediction for low-similarity proteins as they capture meaningful relationships among AA residues that are far away in protein sequence. The proposed prediction method has also been tested to predict structural classes for partially disordered proteins with the reasonable prediction accuracy, which is a more difficult problem comparing to structural class prediction for commonly used benchmark data-sets and has never been done before to the best of our knowledge. In addition, to avoid overfitting with a large number of features, feature selection is applied to select discriminating features that contribute to achieve high prediction accuracy. The selected features have been shown to achieve stable prediction performance across different benchmark data-sets.     

Accurate prediction of protein structural classes using functional domains and predicted secondary structure sequences

Protein structural class prediction is one of the challenging problems in bioinformatics. Previous methods directly based on the similarity of amino acid (AA) sequences have been shown to be insufficient for low-similarity protein data-sets. To improve the prediction accuracy for such low-similarity proteins, different methods have been recently proposed that explore the novel feature sets based on predicted secondary structure propensities. In this paper, we focus on protein structural class prediction using combinations of the novel features including secondary structure propensities as well as functional domain (FD) features extracted from the InterPro signature database. Our comprehensive experimental results based on several benchmark data-sets have shown that the integration of new FD features substantially improves the accuracy of structural class prediction for low-similarity proteins as they capture meaningful relationships among AA residues that are far away in protein sequence. The proposed prediction method has also been tested to predict structural classes for partially disordered proteins with the reasonable prediction accuracy, which is a more difficult problem comparing to structural class prediction for commonly used benchmark data-sets and has never been done before to the best of our knowledge. In addition, to avoid overfitting with a large number of features, feature selection is applied to select discriminating features that contribute to achieve high prediction accuracy. The selected features have been shown to achieve stable prediction performance across different benchmark data-sets.     

New methods for accurate prediction of protein secondary structure.

A primary and a secondary neural network are applied to secondary structure and structural class prediction for a database of 681 non-homologous protein chains. A new method of decoding the outputs of the secondary structure prediction network is used to produce an estimate of the probability of finding each type of secondary structure at every position in the sequence. In addition to providing a reliable estimate of the accuracy of the predictions, this method gives a more accurate Q3 (74.6%) than the cutoff method which is commonly used. Use of these predictions in jury methods improves the Q3 to 74.8%, the best available at present. On a database of 126 proteins commonly used for comparison of prediction methods, the jury predictions are 76.6% accurate. An estimate of the overall Q3 for a given sequence is made by averaging the estimated accuracy of the prediction over all residues in the sequence. As an example, the analysis is applied to the target beta-cryptogein, which was a difficult target for ab initio predictions in the CASP2 study; it shows that the prediction made with the present method (62% of residues correct) is close to the expected accuracy (66%) for this protein. The larger database and use of a new network training protocol also improve structural class prediction accuracy to 86%, relative to 80% obtained previously. Secondary structure content is predicted with accuracy comparable to that obtained with spectroscopic methods, such as vibrational or electronic circular dichroism and Fourier transform infrared spectroscopy.     

SVM-Cabins: prediction of solvent accessibility using accumulation cutoff set and support vector machine

A number of methods for predicting levels of solvent accessibility or accessible surface area (ASA) of amino acid residues in proteins have been developed. These methods either predict regularly spaced states of relative solvent accessibility or an analogue real value indicating relative solvent accessibility. While discrete states of exposure can be easily obtained by post prediction assignment of thresholds to the predicted or computed real values of ASA, the reverse, that is, obtaining a real value from quantized states of predicted ASA, is not straightforward as a two-state prediction in such cases would give a large real valued errors. However, prediction of ASA into larger number of ASA states and then finding a corresponding scheme for real value prediction may be helpful in integrating the two approaches of ASA prediction. We report a novel method of obtaining numerical real values of solvent accessibility, using accumulation cutoff set and support vector machine. This so-called SVM-Cabins method first predicts discrete states of ASA of amino acid residues from their evolutionary profile and then maps the predicted states onto a real valued linear space by simple algebraic methods. Resulting performance of such a rigorous approach using 13-state ASA prediction is at least comparable with the best methods of ASA prediction reported so far. The mean absolute error in this method reaches the best performance of 15.1% on the tested data set of 502 proteins with a coefficient of correlation equal to 0.66. Since, the method starts with the prediction of discrete states of ASA and leads to real value predictions, performance of prediction in binary states and real values are simultaneously optimized.     

Consensus Data Mining (CDM) Protein Secondary Structure Prediction Server: combining GOR V and Fragment Database Mining (FDM)

One of the challenges in protein secondary structure prediction is to overcome the cross-validated 80% prediction accuracy barrier. Here, we propose a novel approach to surpass this barrier. Instead of using a single algorithm that relies on a limited data set for training, we combine two complementary methods having different strengths: Fragment Database Mining (FDM) and GOR V. FDM harnesses the availability of the known protein structures in the Protein Data Bank and provides highly accurate secondary structure predictions when sequentially similar structural fragments are identified. In contrast, the GOR V algorithm is based on information theory, Bayesian statistics, and PSI-BLAST multiple sequence alignments to predict the secondary structure of residues inside a sliding window along a protein chain. A combination of these two different methods benefits from the large number of structures in the PDB and significantly improves the secondary structure prediction accuracy, resulting in Q3 ranging from 67.5 to 93.2%, depending on the availability of highly similar fragments in the Protein Data Bank.     

Improved sequence-based prediction of strand residues

Accurate identification of strand residues aids prediction and analysis of numerous structural and functional aspects of proteins. We propose a sequence-based predictor, BETArPRED, which improves prediction of strand residues and β-strand segments. BETArPRED uses a novel design that accepts strand residues predicted by SSpro and predicts the remaining positions utilizing a logistic regression classifier with nine custom-designed features. These are derived from the primary sequence, the secondary structure (SS) predicted by SSpro, PSIPRED and SPINE, and residue depth as predicted by RDpred. Our features utilize certain local (window-based) patterns in the predicted SS and combine information about the predicted SS and residue depth. BETArPRED is evaluated on 432 sequences that share low identity with the training chains, and on the CASP8 dataset. We compare BETArPRED with seven modern SS predictors, and the top-performing automated structure predictor in CASP8, the ZHANG-server. BETArPRED provides statistically significant improvements over each of the SS predictors; it improves prediction of strand residues and β-strands, and it finds β-strands that were missed by the other methods. When compared with the ZHANG-server, we improve predictions of strand segments and predict more actual strand residues, while the other predictor achieves higher rate of correct strand residue predictions when under-predicting them.     

SVM-based method for protein structural class prediction using secondary structural content and structural information of amino acids

The knowledge collated from the known protein structures has revealed that the proteins are usually folded into the four structural classes: all-α, all-β, α/β and α + β. A number of methods have been proposed to predict the protein's structural class from its primary structure; however, it has been observed that these methods fail or perform poorly in the cases of distantly related sequences. In this paper, we propose a new method for protein structural class prediction using low homology (twilight-zone) protein sequences dataset. Since protein structural class prediction is a typical classification problem, we have developed a Support Vector Machine (SVM)-based method for protein structural class prediction that uses features derived from the predicted secondary structure and predicted burial information of amino acid residues. The examination of different individual as well as feature combinations revealed that the combination of secondary structural content, secondary structural and solvent accessibility state frequencies of amino acids gave rise to the best leave-one-out cross-validation accuracy of ~81% which is comparable to the best accuracy reported in the literature so far.     

Further developments of protein secondary structure prediction using information theory. New parameters and consideration of residue pairs

We have re-evaluated the information used in the Garnier-Osguthorpe-Robson (GOR) method of secondary structure prediction with the currently available database. The framework of information theory provides a means to formulate the influence of local sequence upon the conformation of a given residue, in a rigorous manner. However, the existing database does not allow the evaluation of parameters required for an exact treatment of the problem. The validity of the approximations drawn from the theory is examined. It is shown that the first-level approximation, involving single-residue parameters, is only marginally improved by an increase in the database. The second-level approximation, involving pairs of residues, provides a better model. However, in this case the database is not big enough and this method might lead to parameters with deficiencies. Attention is therefore given to overcoming this lack of data. We have determined the significant pairs and the number of dummy observations necessary to obtain the best result for the prediction. This new version of the GOR method increases the accuracy of prediction by 7%, bringing the amount of residues correctly predicted to 63% for three states and 68 proteins, each protein to be predicted being removed from the database and the parameters derived from the other proteins. If the protein to be predicted is kept in the database the accuracy goes up to 69.7%.     

Combining evolutionary and structural information for local protein structure prediction

We study the effects of various factors in representing and combining evolutionary and structural information for local protein structural prediction based on fragment selection. We prepare databases of fragments from a set of non-redundant protein domains. For each fragment, evolutionary information is derived from homologous sequences and represented as estimated effective counts and frequencies of amino acids (evolutionary frequencies) at each position. Position-specific amino acid preferences called structural frequencies are derived from statistical analysis of discrete local structural environments in database structures. Our method for local structure prediction is based on ranking and selecting database fragments that are most similar to a target fragment. Using secondary structure type as a local structural property, we test our method in a number of settings. The major findings are: (1) the COMPASS-type scoring function for fragment similarity comparison gives better prediction accuracy than three other tested scoring functions for profile-profile comparison. We show that the COMPASS-type scoring function can be derived both in the probabilistic framework and in the framework of statistical potentials. (2) Using the evolutionary frequencies of database fragments gives better prediction accuracy than using structural frequencies. (3) Finer definition of local environments, such as including more side-chain solvent accessibility classes and considering the backbone conformations of neighboring residues, gives increasingly better prediction accuracy using structural frequencies. (4) Combining evolutionary and structural frequencies of database fragments, either in a linear fashion or using a pseudocount mixture formula, results in improvement of prediction accuracy. Combination at the log-odds score level is not as effective as combination at the frequency level. This suggests that there might be better ways of combining sequence and structural information than the commonly used linear combination of log-odds scores. Our method of fragment selection and frequency combination gives reasonable results of secondary structure prediction tested on 56 CASP5 targets (average SOV score 0.77), suggesting that it is a valid method for local protein structure prediction. Mixture of predicted structural frequencies and evolutionary frequencies improve the quality of local profile-to-profile alignment by COMPASS.     

Comparison of the predicted and observed secondary structure of T4 phage lysozyme.

Predictions of the secondary structure of T4 phage lysozyme, made by a number of investigators on the basis of the amino acid sequence, are compared with the structure of the protein determined experimentally by X-ray crystallography. Within the amino terminal half of the molecule the locations of helices predicted by a number of methods agree moderately well with the observed structure, however within the carboxyl half of the molecule the overall agreement is poor. For eleven different helix predictions, the coefficients giving the correlation between prediction and observation range from 0.14 to 0.42. The accuracy of the predictions for both beta-sheet regions and for turns are generally lower than for the helices, and in a number of instances the agreement between prediction and observation is no better than would be expected for a random selection of residues. The structural predictions for T4 phage lysozyme are much less successful than was the case for adenylate kinase (Schulz et al. (1974) Nature 250, 140-142). No one method of prediction is clearly superior to all others, and although empirical predictions based on larger numbers of known protein structure tend to be more accurate than those based on a limited sample, the improvement in accuracy is not dramatic, suggesting that the accuracy of current empirical predictive methods will not be substantially increased simply by the inclusion of more data from additional protein structure determinations.     

Hybrid system for protein secondary structure prediction.

We have developed a hybrid system to predict the secondary structures (alpha-helix, beta-sheet and coil) of proteins and achieved 66.4% accuracy, with correlation coefficients of C(coil) = 0.429, C alpha = 0.470 and C beta = 0.387. This system contains three subsystems ("experts"): a neural network module, a statistical module and a memory-based reasoning module. First, the three experts independently learn the mapping between amino acid sequences and secondary structures from the known protein structures, then a Combiner learns to combine automatically the outputs of the experts to make final predictions. The hybrid system was tested with 107 protein structures through k-way cross-validation. Its performance was better than each expert and all previously reported methods with greater than 0.99 statistical significance. It was observed that for 20% of the residues, all three experts produced the same but wrong predictions. This may suggest an upper bound on the accuracy of secondary structure predictions based on local information from the currently available protein structures, and indicate places where non-local interactions may play a dominant role in conformation. For 64% of the residues, at least two experts were the same and correct, which shows that the Combiner performed better than majority vote. For 77% of the residues, at least one expert was correct, thus there may still be room for improvement in this hybrid approach. Rigorous evaluation procedures were used in testing the hybrid system, and statistical significance measures were developed in analyzing the differences among different methods. When measured in terms of the number of secondary structures (rather than the number of residues) that were predicted correctly, the prediction produced by the hybrid system was also better than those of individual experts.     

NetSurfP-2.0: Improved prediction of protein structural features by integrated deep learning

The ability to predict local structural features of a protein from the primary sequence is of paramount importance for unraveling its function in absence of experimental structural information. Two main factors affect the utility of potential prediction tools: their accuracy must enable extraction of reliable structural information on the proteins of interest, and their runtime must be low to keep pace with sequencing data being generated at a constantly increasing speed. Here, we present NetSurfP-2.0, a novel tool that can predict the most important local structural features with unprecedented accuracy and runtime. NetSurfP-2.0 is sequence-based and uses an architecture composed of convolutional and long short-term memory neural networks trained on solved protein structures. Using a single integrated model, NetSurfP-2.0 predicts solvent accessibility, secondary structure, structural disorder, and backbone dihedral angles for each residue of the input sequences. We assessed the accuracy of NetSurfP-2.0 on several independent test datasets and found it to consistently produce state-of-the-art predictions for each of its output features. We observe a correlation of 80% between predictions and experimental data for solvent accessibility, and a precision of 85% on secondary structure 3-class predictions. In addition to improved accuracy, the processing time has been optimized to allow predicting more than 1000 proteins in less than 2 hours, and complete proteomes in less than 1 day.