Protein Secondary Structure Prediction Methods Research Articles

Combining protein secondary structure prediction models with ensemble methods of optimal complexity

Many sophisticated methods are currently available to perform protein secondary structure prediction. Since they are frequently based on different principles, and different knowledge sources, significant benefits can be expected from combining them. However, the choice of an appropriate combiner appears to be an issue in its own right. The first difficulty to overcome when combining prediction methods is overfitting. This is the reason why we investigate the implementation of Support Vector Machines to perform the task. A family of multi-class SVMs is introduced. Two of these machines are used to combine some of the current best protein secondary structure prediction methods. Their performance is consistently superior to the performance of the ensemble methods traditionally used in the field. They also outperform the decomposition approaches based on bi-class SVMs. Furthermore, initial experimental evidence suggests that their outputs could be processed by the biologist to perform higher-level treatments.

Protein secondary structure prediction based on an improved support vector machines approach.

The prediction of protein secondary structure is an important step in the prediction of protein tertiary structure. A new protein secondary structure prediction method, SVMpsi, was developed to improve the current level of prediction by incorporating new tertiary classifiers and their jury decision system, and the PSI-BLAST PSSM profiles. Additionally, efficient methods to handle unbalanced data and a new optimization strategy for maximizing the Q(3) measure were developed. The SVMpsi produces the highest published Q(3) and SOV94 scores on both the RS126 and CB513 data sets to date. For a new KP480 set, the prediction accuracy of SVMpsi was Q(3) = 78.5% and SOV94 = 82.8%. Moreover, the blind test results for 136 non-redundant protein sequences which do not contain homologues of training data sets were Q(3) = 77.2% and SOV94 = 81.8%. The SVMpsi results in CASP5 illustrate that it is another competitive method to predict protein secondary structure.

Improving Protein Secondary Structure Prediction Methods Using Empirical and Evolutionary Data

Improving Protein Secondary Structure Prediction Methods Using Empirical and Evolutionary Data

Cascaded multiple classifiers for secondary structure prediction.

We describe a new classifier for protein secondary structure prediction that is formed by cascading together different types of classifiers using neural networks and linear discrimination. The new classifier achieves an accuracy of 76.7% (assessed by a rigorous full Jack-knife procedure) on a new nonredundant dataset of 496 nonhomologous sequences (obtained from G.J. Barton and J.A. Cuff). This database was especially designed to train and test protein secondary structure prediction methods, and it uses a more stringent definition of homologous sequence than in previous studies. We show that it is possible to design classifiers that can highly discriminate the three classes (H, E, C) with an accuracy of up to 78% for beta-strands, using only a local window and resampling techniques. This indicates that the importance of long-range interactions for the prediction of beta-strands has been probably previously overestimated.

Improved performance in protein secondary structure prediction by inhomogeneous score combination.

In many fields of pattern recognition, combination has proved efficient to increase the generalization performance of individual prediction methods. Numerous systems have been developed for protein secondary structure prediction, based on different principles. Finding better ensemble methods for this task may thus become crucial. Furthermore, efforts need to be made to help the biologist in the post-processing of the outputs. An ensemble method has been designed to post-process the outputs of discriminant models, in order to obtain an improvement in prediction accuracy while generating class posterior probability estimates. Experimental results establish that it can increase the recognition rate of protein secondary structure prediction methods that provide inhomogeneous scores, even though their individual prediction successes are largely different. This combination thus constitutes a help for the biologist, who can use it confidently on top of any set of prediction methods. Moreover, the resulting estimates can be used in various ways, for instance to determine which areas in the sequence are predicted with a given level of reliability. The prediction is freely available over the Internet on the Network Protein Sequence Analysis (NPS@) WWW server at http://pbil.ibcp.fr/NPSA/npsa_server.ht ml. The source code of the combiner can be obtained on request for academic use.

Identification and application of the concepts important for accurate and reliable protein secondary structure prediction.

A protein secondary structure prediction method from multiply aligned homologous sequences is presented with an overall per residue three-state accuracy of 70.1%. There are two aims: to obtain high accuracy by identification of a set of concepts important for prediction followed by use of linear statistics; and to provide insight into the folding process. The important concepts in secondary structure prediction are identified as: residue conformational propensities, sequence edge effects, moments of hydrophobicity, position of insertions and deletions in aligned homologous sequence, moments of conservation, auto-correlation, residue ratios, secondary structure feedback effects, and filtering. Explicit use of edge effects, moments of conservation, and auto-correlation are new to this paper. The relative importance of the concepts used in prediction was analyzed by stepwise addition of information and examination of weights in the discrimination function. The simple and explicit structure of the prediction allows the method to be reimplemented easily. The accuracy of a prediction is predictable a priori. This permits evaluation of the utility of the prediction: 10% of the chains predicted were identified correctly as having a mean accuracy of > 80%. Existing high-accuracy prediction methods are "black-box" predictors based on complex nonlinear statistics (e.g., neural networks in PHD: Rost & Sander, 1993a). For medium- to short-length chains (> or = 90 residues and < 170 residues), the prediction method is significantly more accurate (P < 0.01) than the PHD algorithm (probably the most commonly used algorithm). In combination with the PHD, an algorithm is formed that is significantly more accurate than either method, with an estimated overall three-state accuracy of 72.4%, the highest accuracy reported for any prediction method.

Improving protein secondary structure prediction with aligned homologous sequences.

Most recent protein secondary structure prediction methods use sequence alignments to improve the prediction quality. We investigate the relationship between the location of secondary structural elements, gaps, and variable residue positions in multiple sequence alignments. We further investigate how these relationships compare with those found in structurally aligned protein families. We show how such associations may be used to improve the quality of prediction of the secondary structure elements, using the Quadratic-Logistic method with profiles. Furthermore, we analyze the extent to which the number of homologous sequences influences the quality of prediction. The analysis of variable residue positions shows that surprisingly, helical regions exhibit greater variability than do coil regions, which are generally thought to be the most common secondary structure elements in loops. However, the correlation between variability and the presence of helices does not significantly improve prediction quality. Gaps are a distinct signal for coil regions. Increasing the coil propensity for those residues occurring in gap regions enhances the overall prediction quality. Prediction accuracy increases initially with the number of homologues, but changes negligibly as the number of homologues exceeds about 14. The alignment quality affects the prediction more than other factors, hence a careful selection and alignment of even a small number of homologues can lead to significant improvements in prediction accuracy.

Protein secondary structure prediction

The past year has seen consolidation of protein secondary structure prediction methods. The advantages of prediction from an aligned family of proteins have been highlighted by several accurate predictions made ‘blind’, before any X-ray or NMR structure was known for the family. New techniques that apply machine learning and discriminant analysis show promise as alternatives to neural networks.

Improvement of protein secondary structure prediction by combination of statistical algorithms and circular dichroism

Three different approaches (propensity curve shifting, hydropathy index evaluation, and iterative attribution/ cancellation of secondary structure) to the use of secondary structure percentages derived from circular dichroism measurements to improve the success rate of a protein secondary structure prediction method, without using decision constants, are described and compared. Propensity-curve shifting appears to be the best-performing approach, bearing an increase of 5.3% in the success rate of single-residue structural prediction when exact information on the secondary structure, obtained by X-ray crystallography, is employed; with information of an accuracy comparable to that obtainable by circular dichroism, the improvement stays between 3.5 and 4.9%, for a three-state prediction. Although developed with circular dichroism in mind, the method can use percentages of secondary structure obtained by any other experimental methodology from which they can be inferred, for instance Raman spectroscopy and infrared spectroscopy.

Analysis of the accuracy and implications of simple methods for predicting the secondary structure of globular proteins

1. (1) Co-operation between a laboratory interested in developing the theory for protein secondary structure prediction methods and a laboratory interested in applying and comparing such methods has led to the development of a simple predictive algorithm. 2. (2) Four-state predictions, in which each residue is unambiguously assigned one conformational state of α-helix, extended chain, reverse turn or coil, predict 49% of residue states correctly (in a sample of 26 proteins) when the overall helix and extended-chain content is not taken into account. 3. (3) When the relative abundances of helix, extended chain, reverse turn and coil observed by X-ray crystallography are taken into account, a single constant for each protein and type of conformation can be used to bias the prediction. When predictions are optimized in this way, 63% of all residue states are unambiguously and correctly assigned. 4. (4) By analysing the nature of the bias required, proteins can be classified into helix-rich types, pleated-sheet-rich types, and so on. It is shown that, if the type of protein can be determined even approximately by circular dichroism, 57% of residue states can be correctly predicted without taking into account the X-ray structure. Further, comparable predictions can be obtained if, instead of circular dichroism, preliminary predictions are made to assess the protein type. 5. (5) It is emphasized that the numbers quoted here depend on the method used to assess accuracy, and the algorithm is shown to be at least as good as, and usually superior to, the reported prediction methods assessed in the same way. 6. (6) Ways of further enhancing predictions by the use of additional information from hydrophobic triplets and homologous sequences are also explored. Hydro-phobic triplet information does not significantly improve predictive power and it is concluded that this information is used by proteins in the next stage of folding. On the other hand, the use of homologous sequences appears to be very promising. 7. (7) The implication of these results in protein folding is discussed.

An assessment of protein secondary structure prediction methods based on amino acid sequence

Five of the several secondary structure prediction methods based on protein amino acid sequence have been computerized, allowing the calculation of joint prediction histograms which have been shown to be superior to any individual prediction. The known structures of about 40 proteins experimentally determined by X-ray crystallography are compared with the predictions resulting from calculated histograms. The accuracy of the predictions for helices is generally much better than for both β-sheet regions and for turns. The overall agreement between prediction and observation within the amino terminal half of the protein molecules is clearly superior to that for the carboxyl half, suggesting an amino nucleating core. Predictions for smaller proteins and thermally stable proteins are generally good, indicating the sensitivity of the methods to short-range but not long-range interactions. In less than half the cases tested were the predictions useful; there was no way of knowing ahead of time if a favorable prediction would result. Given the lack of dramatic improvement with an increase in data base for the schemes and the generally poor agreement factors, it appears that a perfect predictive algorithm must include a consideration of energy minimization, thermalization, and long-range interactions. Extreme caution is suggested in applying present prediction routines to unknown protein structures.