Support vector machines for prediction and analysis of beta and gamma-turns in proteins

Tight turns have long been recognized as one of the three important features of proteins, together with alpha-helix and beta-sheet. Tight turns play an important role in globular proteins from both the structural and functional points of view. More than 90% tight turns are beta-turns and most of the rest are gamma-turns. Analysis and prediction of beta-turns and gamma-turns is very useful for design of new molecules such as drugs, pesticides, and antigens. In this paper we investigated two aspects of applying support vector machine (SVM), a promising machine learning method for bioinformatics, to prediction and analysis of beta-turns and gamma-turns. First, we developed two SVM-based methods, called BTSVM and GTSVM, which predict beta-turns and gamma-turns in a protein from its sequence. When compared with other methods, BTSVM has a superior performance and GTSVM is competitive. Second, we used SVMs with a linear kernel to estimate the support of amino acids for the formation of beta-turns and gamma-turns depending on their position in a protein. Our analysis results are more comprehensive and easier to use than the previous results in designing turns in proteins.     

A neural network method for prediction of beta-turn types in proteins using evolutionary information

MOTIVATION: The prediction of beta-turns is an important element of protein secondary structure prediction. Recently, a highly accurate neural network based method Betatpred2 has been developed for predicting beta-turns in proteins using position-specific scoring matrices (PSSM) generated by PSI-BLAST and secondary structure information predicted by PSIPRED. However, the major limitation of Betatpred2 is that it predicts only beta-turn and non-beta-turn residues and does not provide any information of different beta-turn types. Thus, there is a need to predict beta-turn types using an approach based on multiple sequence alignment, which will be useful in overall tertiary structure prediction. RESULTS: In the present work, a method has been developed for the prediction of beta-turn types I, II, IV and VIII. For each turn type, two consecutive feed-forward back-propagation networks with a single hidden layer have been used where the first sequence-to-structure network has been trained on single sequences as well as on PSI-BLAST PSSM. The output from the first network along with PSIPRED predicted secondary structure has been used as input for the second-level structure-to-structure network. The networks have been trained and tested on a non-homologous dataset of 426 proteins chains by 7-fold cross-validation. It has been observed that the prediction performance for each turn type is improved significantly by using multiple sequence alignment. The performance has been further improved by using a second level structure-to-structure network and PSIPRED predicted secondary structure information. It has been observed that Type I and II beta-turns have better prediction performance than Type IV and VIII beta-turns. The final network yields an overall accuracy of 74.5, 93.5, 67.9 and 96.5% with MCC values of 0.29, 0.29, 0.23 and 0.02 for Type I, II, IV and VIII beta-turns, respectively, and is better than random prediction. AVAILABILITY: A web server for prediction of beta-turn types I, II, IV and VIII based on above approach is available at http://www.imtech.res.in/raghava/betaturns/ and http://bioinformatics.uams.edu/mirror/betaturns/ (mirror site).     

An evaluation of beta-turn prediction methods

MOTIVATION: beta-turn is an important element of protein structure. In the past three decades, numerous beta-turn prediction methods have been developed based on various strategies. For a detailed discussion about the importance of beta-turns and a systematic introduction of the existing prediction algorithms for beta-turns and their types, please see a recent review (Chou, Analytical Biochemistry, 286, 1-16, 2000). However at present, it is still difficult to say which method is better than the other. This is because of the fact that these methods were developed on different sets of data. Thus, it is important to evaluate the performance of beta-turn prediction methods. RESULTS: We have evaluated the performance of six methods of beta-turn prediction. All the methods have been tested on a set of 426 non-homologous protein chains. It has been observed that the performance of the neural network based method, BTPRED, is significantly better than the statistical methods. One of the reasons for its better performance is that it utilizes the predicted secondary structure information. We have also trained, tested and evaluated the performance of all methods except BTPRED and GORBTURN, on new data set using a 7-fold cross-validation technique. There is a significant improvement in performance of all the methods when secondary structure information is incorporated. Moreover, after incorporating secondary structure information, the Sequence Coupled Model has yielded better results in predicting beta-turns as compared with other methods. In this study, both threshold dependent and independent (ROC) measures have been used for evaluation.     

Protein beta-turn prediction using nearest-neighbor method

MOTIVATION: With the emerging success of protein secondary structure prediction through the applications of various statistical and machine learning techniques, similar techniques have been applied to protein beta-turn prediction. In this study, we perform protein beta-turn prediction using a k-nearest neighbor method, which is combined with a filter that uses predicted protein secondary structure information. Traditional beta-turn prediction from k-nearest neighbor method is modified to account for the unbalanced ratio of the natural occurrence of beta-turns and non-beta-turns. RESULTS: Our prediction scheme is tested on a set of 426 non-homologous protein sequences. The prediction scheme consists of two stages: k-nearest neighbor method stage and filtering stage. Variations of the k-nearest neighbor method were used to take property of beta-turns into consideration. Our filtering method uses beta-turn/non-beta-turn estimates from the k-nearest neighbor method stage and predicted protein secondary structure information from PSI-PRED in order to get new beta-turn/non-beta-turn estimate. Our result is compared with the previously best known beta-turn prediction method on the dataset of 426 non-homologous protein sequences and is shown to give slightly superior performance at significantly lower computational complexity. AVAILABILITY: Contact the author for information on the source code of the programs used.     

Analysis and prediction of the different types of beta-turn in proteins

beta-Turns have been extracted from 59 non-identical proteins (resolution 2 A) using the standard criterion that the distance between C alpha (i) and C alpha (i + 3) is less than 7 A (1 A = 0.1 nm). The beta-turns have been classified, using phi, psi angles, into seven conventional turn types (I, I', II, II', IV, VIa, VIb) and a new class of beta-turn, designated type VIII, in which the central residues (i + 1, i + 2) adopt an alpha R beta conformation. Most beta-turn types are found in various topological environments, with the exception of I' and II' beta-turns, where 83% and 50%, respectively, are found in beta-hairpins. Sufficient data have been gathered to enable, for the first time, the separate statistical analysis of type I and II beta-turns. The two turn types have been shown to be strikingly different in their sequence preferences. Type I turns favour Asp, Asn, Ser and Cys at i; Asp, Ser, Thr and Pro at i + 1; Asp, Ser, Asn and Arg at i + 2; Gly, Trp and Met at i + 3, whilst type II turns prefer Pro at i + 1; Gly and Asn at i + 2; Gln and Arg at i + 3. These preferences have been explained by the specific side-chain interactions observed within the X-ray structures. The positional trends for type I and II beta-turns have been incorporated into the simple empirical predictive algorithm originally developed by P.N. Lewis et al. The program has improved the positional prediction of beta-turns, and has enhanced and extended the method by predicting the type of beta-turn. Since the observed preferences reflect local interactions these predictions are applicable not only to proteins, but also to peptides, many of which are thought to contain beta-turns.     

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).     

Prediction of pi-turns in proteins using PSI-BLAST profiles and secondary structure information

Due to the structural and functional importance of tight turns, some methods have been proposed to predict gamma-turns, beta-turns, and alpha-turns in proteins. In the past, studies of pi-turns were made, but not a single prediction approach has been developed so far. It will be useful to develop a method for identifying pi-turns in a protein sequence. In this paper, the support vector machine (SVM) method has been introduced to predict pi-turns from the amino acid sequence. The training and testing of this approach is performed with a newly collected data set of 640 non-homologous protein chains containing 1931 pi-turns. Different sequence encoding schemes have been explored in order to investigate their effects on the prediction performance. With multiple sequence alignment and predicted secondary structure, the final SVM model yields a Matthews correlation coefficient (MCC) of 0.556 by a 7-fold cross-validation. A web server implementing the prediction method is available at the following URL: http://210.42.106.80/piturn/.     

PEPstr: a de novo method for tertiary structure prediction of small bioactive peptides

Among secondary structure elements, beta-turns are ubiquitous and major feature of bioactive peptides. We analyzed 77 biologically active peptides with length varying from 9 to 20 residues. Out of 77 peptides, 58 peptides were found to contain at least one beta-turn. Further, at the residue level, 34.9% of total peptide residues were found to be in beta-turns, higher than the number of helical (32.3%) and beta-sheet residues (6.9%). So, we utilized the predicted beta-turns information to develop an improved method for predicting the three-dimensional (3D) structure of small peptides. In principle, we built four different structural models for each peptide. The first 'model I' was built by assigning all the peptide residues an extended conformation (phi = Psi = 180 degrees ). Second 'model II' was built using the information of regular secondary structures (helices, beta-strands and coil) predicted from PSIPRED. In third 'model III', secondary structure information including beta-turn types predicted from BetaTurns method was used. The fourth 'model IV' had main-chain phi, Psi angles of model III and side chain angles assigned using standard Dunbrack backbone dependent rotamer library. These models were further refined using AMBER package and the resultant C(alpha) rmsd values were calculated. It was found that adding the beta-turns to the regular secondary structures greatly reduces the rmsd values both before and after the energy minimization. Hence, the results indicate that regular and irregular secondary structures, particularly beta-turns information can provide valuable and vital information in the tertiary structure prediction of small bioactive peptides. Based on the above study, a web server PEPstr (http://www.imtech.res.in/raghava/pepstr/) was developed for predicting the tertiary structure of small bioactive peptides.     

Prediction of beta-turns in proteins from multiple alignment using neural network

A neural network-based method has been developed for the prediction of beta-turns in proteins by using multiple sequence alignment. Two feed-forward back-propagation networks with a single hidden layer are used where the first-sequence structure network is trained with the multiple sequence alignment in the form of PSI-BLAST-generated position-specific scoring matrices. The initial predictions from the first network and PSIPRED-predicted secondary structure are used as input to the second structure-structure network to refine the predictions obtained from the first net. A significant improvement in prediction accuracy has been achieved by using evolutionary information contained in the multiple sequence alignment. The final network yields an overall prediction accuracy of 75.5% when tested by sevenfold cross-validation on a set of 426 nonhomologous protein chains. The corresponding Q(pred), Q(obs), and Matthews correlation coefficient values are 49.8%, 72.3%, and 0.43, respectively, and are the best among all the previously published beta-turn prediction methods. The Web server BetaTPred2 (http://www.imtech.res.in/raghava/betatpred2/) has been developed based on this approach.     

Studies on hydrogen bonds. Part V--Hydrogen bonding in energy minimization studies of peptides

An energy term, representing the N-H...O type of hydrogen bond, which is a function of the hydrogen bond length (R) and angle (theta) has been introduced in an energy minimization program, taking into consideration its interpolation with the non-bonded energy for borderline values of R and theta. The details of the mathematical formulation of the derivatives of the hydrogen bond function as applicable to the energy minimization have been given. The minimization technique has been applied to hydrogen bonded two and three linked peptide units (gamma-turns and beta-turns), and having Gly, Ala and Pro side chains. Some of the conformational highlights of the resulting minimum energy conformations are a) the occurrence of the expected 4----1 hydrogen bond in all of the burn-turn tripeptide sequences and b) the presence of an additional 3----1 hydrogen bond in some of the type I and II tripeptides with the hydrogen bonding scheme in such type I beta-turns occurring in a bifurcated form. These and other conformational features have been discussed in the light of experimental evidence and theoretical predictions of other workers.     

Improved method for predicting beta-turn using support vector machine

MOTIVATION: Numerous methods for predicting beta-turns in proteins have been developed based on various computational schemes. Here, we introduce a new method of beta-turn prediction that uses the support vector machine (SVM) algorithm together with predicted secondary structure information. Various parameters from the SVM have been adjusted to achieve optimal prediction performance. RESULTS: The SVM method achieved excellent performance as measured by the Matthews correlation coefficient (MCC = 0.45) using a 7-fold cross validation on a database of 426 non-homologous protein chains. To our best knowledge, this MCC value is the highest achieved so far for predicting beta-turn. The overall prediction accuracy Qtotal was 77.3%, which is the best among the existing prediction methods. Among its unique attractive features, the present SVM method avoids overtraining and compresses information and provides a predicted reliability index.     

Support vector machine for predicting alpha-turn types

Tight turns play an important role in globular proteins from both the structural and functional points of view. Of tight turns, beta-turns and gamma-turns have been extensively studied, but alpha-turns were little investigated. Recently, a systematic search for alpha-turns classified alpha-turns into nine different types according to their backbone trajectory features. In this paper, Support Vector Machines (SVMs), a new machine learning method, is proposed for predicting the alpha-turn types in proteins. The high rates of correct prediction imply that that the formation of different alpha-turn types is evidently correlated with the sequence of a pentapeptide, and hence can be approximately predicted based on the sequence information of the pentapeptide alone, although the incorporation of its interaction with the other part of a protein, the so-called "long distance interaction", will further improve the prediction quality.     

Prediction of beta-turns in proteins using neural networks

The use of neural networks to improve empirical secondary structure prediction is explored with regard to the identification of the position and conformational class of beta-turns, a four-residue chain reversal. Recently an algorithm was developed for beta-turn predictions based on the empirical approach of Chou and Fasman using different parameters for three classes (I, II and non-specific) of beta-turns. In this paper, using the same data, an alternative approach to derive an empirical prediction method is used based on neural networks which is a general learning algorithm extensively used in artificial intelligence. Thus the results of the two approaches can be compared. The most severe test of prediction accuracy is the percentage of turn predictions that are correct and the neural network gives an overall improvement from 20.6% to 26.0%. The proportion of correctly predicted residues is 71%, compared to a chance level of about 58%. Thus neural networks provide a method of obtaining more accurate predictions from empirical data than a simpler method of deriving propensities.     

The heat denaturation of bacterial ribonuclease studied by infrared spectroscopy]

The thermal denaturation of bacterial ribonuclease in the interval of pH 2.5-7.0 has been investigated by means of infra-red spectroscopy method. The protein melting for pH 2.5 begins at the temperature 25 degrees C and is accompanied by secondary protein structure reconstruction, partially destroying native beta-structure and leading to new denatured conformation appearance of different types of beta-turns. Spectral changes for pH 3.5 and 7.0 are significantly less in the same frequency areas. At the temperature more than 50 degrees C protein aggregation takes place with inter-molecule-beta-form formation.     

Prediction of beta-turns from amino acid sequences using the residue-coupled model.

We evaluated the prediction of beta-turns from amino acid sequences using the residue-coupled model with an enlarged representative protein data set selected from the Protein Data Bank. Our results show that the probability values derived from a data set comprising 425 protein chains yielded an overall beta-turn prediction accuracy 68.74%, compared with 94.7% reported earlier on a data set of 30 proteins using the same method. However, we noted that the overall beta-turn prediction accuracy using probability values derived from the 30-protein data set reduces to 40.74% when tested on the data set comprising 425 protein chains. In contrast, using probability values derived from the 425 data set used in this analysis, the overall beta-turn prediction accuracy yielded consistent results when tested on either the 30-protein data set (64.62%) used earlier or a more recent representative data set comprising 619 protein chains (64.66%) or on a jackknife data set comprising 476 representative protein chains (63.38%). We therefore recommend the use of probability values derived from the 425 representative protein chains data set reported here, which gives more realistic and consistent predictions of beta-turns from amino acid sequences.     

Neural network prediction of 3(10)-helices in proteins

Secondary structure prediction from the primary sequence of a protein is fundamental to understanding its structure and folding properties. Although several prediction methodologies are in vogue, their performances are far from being completely satisfactory. Among these, non-linear neural networks have been shown to be relatively effective, especially for predicting beta-turns, where dominant interactions are local, arising from four sequence-contiguous residues. Most 3(10)-helices in proteins are also short, comprising of three sequence-contiguous residues and two capping residues. In order to understand the extent of local interactions in these 3(10)-helices, we have applied a neural network model with varying window size to predict 3(10)-helices in proteins. We found the prediction accuracy of 3(10)-helices (approximately 14%), as judged by the Matthew's Correlation Coefficient, to be less than that of beta-turns (approximately 20%). The optimal window size for the prediction of 3(10)-helices was about 9 residues. The significance and implications of these results in understanding the occurrence of 3(10)-helices and preferences of amino acid residues in 3(10)-helices are discussed.     

Protein beta-turn assignments

A classical way to analyze protein 3D structures or models is to investigate their secondary structures. Their predictions are also widely used as a help to build new 3D models. Thus, hundreds of prediction methods have been proposed. Nonetheless before predicting, secondary structure assignment is required even if not trivial. Therefore numerous but diverging assignment methods have been developed. Beta-turns constitute the third most important secondary structures. However, no analysis to compare the beta-turn distributions according to different secondary structure assignment methods has ever been done. We propose in this paper to analyze and evaluate the results of such a comparison. We highlight some important divergence that could have important consequence for the analysis and prediction of beta-turns.     

BetaTPred: prediction of beta-TURNS in a protein using statistical algorithms

MOTIVATION: beta-turns play an important role from a structural and functional point of view. beta-turns are the most common type of non-repetitive structures in proteins and comprise on average, 25% of the residues. In the past numerous methods have been developed to predict beta-turns in a protein. Most of these prediction methods are based on statistical approaches. In order to utilize the full potential of these methods, there is a need to develop a web server. RESULTS: This paper describes a web server called BetaTPred, developed for predicting beta-TURNS in a protein from its amino acid sequence. BetaTPred allows the user to predict turns in a protein using existing statistical algorithms. It also allows to predict different types of beta-TURNS e.g. type I, I', II, II', VI, VIII and non-specific. This server assists the users in predicting the consensus beta-TURNS in a protein. AVAILABILITY: The server is accessible from http://imtech.res.in/raghava/betatpred/     

Mimicry of beta II'-turns of proteins in cyclic pentapeptides with one and without D-amino acids.

The solution structure of eight cyclic pentapeptides has been determined by two-dimensional 1H-NMR spectroscopy combined with spectra simulations and restrained molecular dynamic simulations. Six of the cyclic pentapeptides were derived from the C-terminal cholecystokinin fragment CCK-4 enlarged with Asp1 resulting in the sequence (Asp-Trp-Met-Asp-Phe), one L-amino acid after the other was substituted by its D-analog. In addition, two peptides, including an all-L-amino-acid-containing cyclic pentapeptide, cyclo(Asp-Phe-Lys-Ala-Thr) and cyclo(Asp-Phe-Lys-Ala-D-Thr) were investigated. All D-amino-acid-containing peptides show beta II'-turn conformations with the D-amino acid in the i + 1 position, excepting the D-aspartic-acid-containing peptides. These two peptides are characterized by the lack of beta-turns at pH values less than 4, suggesting that D-aspartic acid in the full-protonized state avoids the formation of beta-turns in these compounds. At pH values greater than 5, a conformational change into the beta II'-turn conformation was also observed for these peptides. Conformations without beta-turns are expected for cyclic all-L pentapeptides, but both cyclo(Asp-Phe-Lys-Ala-Thr) and the D-Thr analog cyclo(Asp-Phe-Lys-Ala-D-Thr) exhibit beta II'-turn conformations around Thr-Asp and D-Thr-Asp. Thus cyclic all-L pentapeptides and those with one D-amino acid are able to form similar structures preferably with a beta II'-turn. The beta-turn formation in cyclic pentapeptides containing a D-aspartic acid is dependent on the ionization state. The relevance of the work to the design of beta'-turn mimetics is discussed.