Combining prediction of secondary structure and solvent accessibility in proteins

Owing to the use of evolutionary information and advanced machine learning protocols, secondary structures of amino acid residues in proteins can be predicted from the primary sequence with more than 75% per-residue accuracy for the 3-state (i.e., helix, beta-strand, and coil) classification problem. In this work we investigate whether further progress may be achieved by incorporating the relative solvent accessibility (RSA) of an amino acid residue as a fingerprint of the overall topology of the protein. Toward that goal, we developed a novel method for secondary structure prediction that uses predicted RSA in addition to attributes derived from evolutionary profiles. Our general approach follows the 2-stage protocol of Rost and Sander, with a number of Elman-type recurrent neural networks (NNs) combined into a consensus predictor. The RSA is predicted using our recently developed regression-based method that provides real-valued RSA, with the overall correlation coefficients between the actual and predicted RSA of about 0.66 in rigorous tests on independent control sets. Using the predicted RSA, we were able to improve the performance of our secondary structure prediction by up to 1.4% and achieved the overall per-residue accuracy between 77.0% and 78.4% for the 3-state classification problem on different control sets comprising, together, 603 proteins without homology to proteins included in the training. The effects of including solvent accessibility depend on the quality of RSA prediction. In the limit of perfect prediction (i.e., when using the actual RSA values derived from known protein structures), the accuracy of secondary structure prediction increases by up to 4%. We also observed that projecting real-valued RSA into 2 discrete classes with the commonly used threshold of 25% RSA decreases the classification accuracy for secondary structure prediction. While the level of improvement of secondary structure prediction may be different for prediction protocols that implicitly account for RSA in other ways, we conclude that an increase in the 3-state classification accuracy may be achieved when combining RSA with a state-of-the-art protocol utilizing evolutionary profiles. The new method is available through a Web server at http://sable.cchmc.org.     

Statistical torsion angle potential energy functions for protein structure modeling: A bicubic interpolation approach

A set of grid type knowledge‐based energy functions is introduced for ?–χ₁, ψ–χ₁, ?–ψ, and χ₁–χ₂ torsion angle combinations. Boltzmann distribution is assumed for the torsion angle populations from protein X‐ray structures, and the functions are named as statistical torsion angle potential energy functions. The grid points around periodic boundaries are duplicated to force periodicity, and the remedy relieves the derivative discontinuity problem. The devised functions rapidly improve the quality of model structures. The potential bias in the functions and the usefulness of additional secondary structure information are also investigated. The proposed guiding functions are expected to facilitate protein structure modeling, such as protein structure prediction, protein design, and structure refinement. Proteins 2013. Proteins 2013; 81:1156–1165. © 2013 Wiley Periodicals, Inc.     

Predicting protein secondary structure and solvent accessibility with an improved multiple linear regression method

We have improved the multiple linear regression (MLR) algorithm for protein secondary structure prediction by combining it with the evolutionary information provided by multiple sequence alignment of PSI-BLAST. On the CB513 dataset, the three states average overall per-residue accuracy, Q(3), reached 76.4%, while segment overlap accuracy, SOV99, reached 73.2%, using a rigorous jackknife procedure and the strictest reduction of eight states DSSP definition to three states. This represents an improvement of approximately 5% on overall per-residue accuracy compared with previous work. The relative solvent accessibility prediction also benefited from this combination of methods. The system achieved 77.7% average jackknifed accuracy for two states prediction based on a 25% relative solvent accessibility mode, with a Mathews' correlation coefficient of 0.548. The improved MLR secondary structure and relative solvent accessibility prediction server is available at http://spg.biosci.tsinghua.edu.cn/.     

Protein sequence randomization: efficient estimation of protein stability using knowledge-based potentials

Modifications of the amino acid sequence generally affect protein stability. Here, we use knowledge-based potentials to estimate the stability of protein structures under sequence variation. Calculations on a variety of protein scaffolds result in a clear distinction of known mutable regions from arbitrarily chosen control patches. For example, randomly changing the sequence of an antibody paratope yields a significantly lower number of destabilized mutants as compared to the randomization of comparable regions on the protein surface. The technique is computationally efficient and can be used to screen protein structures for regions that are amenable to molecular tinkering by preserving the stability of the mutated proteins.     

Real value prediction of solvent accessibility in proteins using multiple sequence alignment and secondary structure

The present study is an attempt to develop a neural network-based method for predicting the real value of solvent accessibility from the sequence using evolutionary information in the form of multiple sequence alignment. In this method, two feed-forward networks with a single hidden layer have been trained with standard back-propagation as a learning algorithm. The Pearson's correlation coefficient increases from 0.53 to 0.63, and mean absolute error decreases from 18.2 to 16% when multiple-sequence alignment obtained from PSI-BLAST is used as input instead of a single sequence. The performance of the method further improves from a correlation coefficient of 0.63 to 0.67 when secondary structure information predicted by PSIPRED is incorporated in the prediction. The final network yields a mean absolute error value of 15.2% between the experimental and predicted values, when tested on two different nonhomologous and nonredundant datasets of varying sizes. The method consists of two steps: (1) in the first step, a sequence-to-structure network is trained with the multiple alignment profiles in the form of PSI-BLAST-generated position-specific scoring matrices, and (2) in the second step, the output obtained from the first network and PSIPRED-predicted secondary structure information is used as an input to the second structure-to-structure network. Based on the present study, a server SARpred (http://www.imtech.res.in/raghava/sarpred/) has been developed that predicts the real value of solvent accessibility of residues for a given protein sequence. We have also evaluated the performance of SARpred on 47 proteins used in CASP6 and achieved a correlation coefficient of 0.68 and a MAE of 15.9% between predicted and observed values.     

Information‐theoretic analysis of the reference state in contact potentials used for protein structure prediction

Using information‐theoretic concepts, we examine the role of the reference state, a crucial component of empirical potential functions, in protein fold recognition. We derive an information‐based connection between the probability distribution functions of the reference state and those that characterize the decoy set used in threading. In examining commonly used contact reference states, we find that the quasi‐chemical approximation is informatically superior to other variant models designed to include characteristics of real protein chains, such as finite length and variable amino acid composition from protein to protein. We observe that in these variant models, the total divergence, the operative function that quantifies discrimination, decreases along with threading performance. We find that any amount of nativeness encoded in the reference state model does not significantly improve threading performance. A promising avenue for the development of better potentials is suggested by our information‐theoretic analysis of the action of contact potentials on individual protein sequences. Our results show that contact potentials perform better when the compositional properties of the data set used to derive the score function probabilities are similar to the properties of the sequence of interest. Results also suggest to use only sequences of similar composition in deriving contact potentials, to tailor the contact potential specifically for a test sequence. Proteins 2010. © 2009 Wiley‐Liss, Inc.     

Understanding the role of Arg96 in structure and stability of green fluorescent protein

Arg96 is a highly conservative residue known to catalyze spontaneous green fluorescent protein (GFP) chromophore biosynthesis. To understand a role of Arg96 in conformational stability and structural behavior of EGFP, the properties of a series of the EGFP mutants bearing substitutions at this position were studied using circular dichroism, steady state fluorescence spectroscopy, fluorescence lifetime, kinetics and equilibrium unfolding analysis, and acrylamide-induced fluorescence quenching. During the protein production and purification, high yield was achieved for EGFP/Arg96Cys variant, whereas EGFP/Arg96Ser and EGFP/Arg96Ala were characterized by essentially lower yields and no protein was produced when Arg96 was substituted by Gly. We have also shown that only EGFP/Arg96Cys possessed relatively fast chromophore maturation, whereas it took EGFP/Arg96Ser and EGFP/Arg96Ala about a year to develop a noticeable green fluorescence. The intensity of the characteristic green fluorescence measured for the EGFP/Arg96Cys and EGFP/Arg96Ser (or EGFP/Arg96Ala) was 5- and 50-times lower than that of the nonmodified EGFP. Intriguingly, EGFP/Arg96Cys was shown to be more stable than EGFP toward the GdmCl-induced unfolding both in kinetics and in the quasi-equilibrium experiments. In comparison with EGFP, tryptophan residues of EGFP/Arg96Cys were more accessible to the solvent. These data taken together suggest that besides established earlier crucial catalytic role, Arg96 is important for the overall folding and conformational stability of GFP.     

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.     

Simultaneous prediction of protein secondary structure and transmembrane spans

Prediction of transmembrane spans and secondary structure from the protein sequence is generally the first step in the structural characterization of (membrane) proteins. Preference of a stretch of amino acids in a protein to form secondary structure and being placed in the membrane are correlated. Nevertheless, current methods predict either secondary structure or individual transmembrane states. We introduce a method that simultaneously predicts the secondary structure and transmembrane spans from the protein sequence. This approach not only eliminates the necessity to create a consensus prediction from possibly contradicting outputs of several predictors but bears the potential to predict conformational switches, i.e., sequence regions that have a high probability to change for example from a coil conformation in solution to an α‐helical transmembrane state. An artificial neural network was trained on databases of 177 membrane proteins and 6048 soluble proteins. The output is a 3 × 3 dimensional probability matrix for each residue in the sequence that combines three secondary structure types (helix, strand, coil) and three environment types (membrane core, interface, solution). The prediction accuracies are 70.3% for nine possible states, 73.2% for three‐state secondary structure prediction, and 94.8% for three‐state transmembrane span prediction. These accuracies are comparable to state‐of‐the‐art predictors of secondary structure (e.g., Psipred) or transmembrane placement (e.g., OCTOPUS). The method is available as web server and for download at www.meilerlab.org . Proteins 2013; 81:1127–1140. © 2013 Wiley Periodicals, Inc.     

Structural refinement of protein segments containing secondary structure elements: Local sampling, knowledge-based potentials, and clustering

In this article, we present an iterative, modular optimization (IMO) protocol for the local structure refinement of protein segments containing secondary structure elements (SSEs). The protocol is based on three modules: a torsion-space local sampling algorithm, a knowledge-based potential, and a conformational clustering algorithm. Alternative methods are tested for each module in the protocol. For each segment, random initial conformations were constructed by perturbing the native dihedral angles of loops (and SSEs) of the segment to be refined while keeping the protein body fixed. Two refinement procedures based on molecular mechanics force fields - using either energy minimization or molecular dynamics - were also tested but were found to be less successful than the IMO protocol. We found that DFIRE is a particularly effective knowledge-based potential and that clustering algorithms that are biased by the DFIRE energies improve the overall results. Results were further improved by adding an energy minimization step to the conformations generated with the IMO procedure, suggesting that hybrid strategies that combine both knowledge-based and physical effective energy functions may prove to be particularly effective in future applications.     

Prediction and evolutionary information analysis of protein solvent accessibility using multiple linear regression

A multiple linear regression method was applied to predict real values of solvent accessibility from the sequence and evolutionary information. This method allowed us to obtain coefficients of regression and correlation between the occurrence of an amino-acid residue at a specific target and its sequence neighbor positions on the one hand, and the solvent accessibility of that residue on the other. Our linear regression model based on sequence information and evolutionary models was found to predict residue accessibility with 18.9% and 16.2% mean absolute error respectively, which is better than or comparable to the best available methods. A correlation matrix for several neighbor positions to examine the role of evolutionary information at these positions has been developed and analyzed. As expected, the effective frequency of hydrophobic residues at target positions shows a strong negative correlation with solvent accessibility, whereas the reverse is true for charged and polar residues. The correlation of solvent accessibility with effective frequencies at neighboring positions falls abruptly with distance from target residues. Longer protein chains have been found to be more accurately predicted than their smaller counterparts.     

A novel method for protein secondary structure prediction using dual-layer SVM and profiles

A high-performance method was developed for protein secondary structure prediction based on the dual-layer support vector machine (SVM) and position-specific scoring matrices (PSSMs). SVM is a new machine learning technology that has been successfully applied in solving problems in the field of bioinformatics. The SVM's performance is usually better than that of traditional machine learning approaches. The performance was further improved by combining PSSM profiles with the SVM analysis. The PSSMs were generated from PSI-BLAST profiles, which contain important evolution information. The final prediction results were generated from the second SVM layer output. On the CB513 data set, the three-state overall per-residue accuracy, Q3, reached 75.2%, while segment overlap (SOV) accuracy increased to 80.0%. On the CB396 data set, the Q3 of our method reached 74.0% and the SOV reached 78.1%. A web server utilizing the method has been constructed and is available at http://www.bioinfo.tsinghua.edu.cn/pmsvm.     

Ab initio protein structure assembly using continuous structure fragments and optimized knowledge-based force field

Ab initio protein folding is one of the major unsolved problems in computational biology owing to the difficulties in force field design and conformational search. We developed a novel program, QUARK, for template-free protein structure prediction. Query sequences are first broken into fragments of 1-20 residues where multiple fragment structures are retrieved at each position from unrelated experimental structures. Full-length structure models are then assembled from fragments using replica-exchange Monte Carlo simulations, which are guided by a composite knowledge-based force field. A number of novel energy terms and Monte Carlo movements are introduced and the particular contributions to enhancing the efficiency of both force field and search engine are analyzed in detail. QUARK prediction procedure is depicted and tested on the structure modeling of 145 nonhomologous proteins. Although no global templates are used and all fragments from experimental structures with template modeling score >0.5 are excluded, QUARK can successfully construct 3D models of correct folds in one-third cases of short proteins up to 100 residues. In the ninth community-wide Critical Assessment of protein Structure Prediction experiment, QUARK server outperformed the second and third best servers by 18 and 47% based on the cumulative Z-score of global distance test-total scores in the FM category. Although ab initio protein folding remains a significant challenge, these data demonstrate new progress toward the solution of the most important problem in the field.     

Multibody coarse-grained potentials for native structure recognition and quality assessment of protein models

Multibody potentials have been of much interest recently because they take into account three dimensional interactions related to residue packing and capture the cooperativity of these interactions in protein structures. Our goal was to combine long range multibody potentials and short range potentials to improve recognition of native structure among misfolded decoys. We optimized the weights for four-body nonsequential, four-body sequential, and short range potentials to obtain optimal model ranking results for threading and have compared these data against results obtained with other potentials (26 different coarse-grained potentials from the Potentials 'R'Us web server have been used). Our optimized multibody potentials outperform all other contact potentials in the recognition of the native structure among decoys, both for models from homology template-based modeling and from template-free modeling in CASP8 decoy sets. We have compared the results obtained for this optimized coarse-grained potentials, where each residue is represented by a single point, with results obtained by using the DFIRE potential, which takes into account atomic level information of proteins. We found that for all proteins larger than 80 amino acids our optimized coarse-grained potentials yield results comparable to those obtained with the atomic DFIRE potential.     

Positioning of anchor groups in protein loop prediction: the importance of solvent accessibility and secondary structure elements

The prediction of loop regions in the process of protein structure prediction by homology is still an unsolved problem. In an earlier publication, we could show that the correct placement of the amino acids serving as an anchor group to be connected by a loop fragment with a predicted geometry is a highly important step and an essential requirement within the process (Lessel and Schomburg, Proteins 1999; 37:56-64). In this article, we present an analysis of the quality of possible loop predictions with respect to gap length, fragment length, amino acid type, secondary structure, and solvent accessibility. For 550 insertions and 544 deletions, we test all possible positions for anchor groups with an inserted loop of a length between 3 and 12 amino acids. We could show that approximately 80% of the indel regions could be predicted within 1.5 A RMSD from a knowledge-based loop data base if criteria for the correct localization of anchor groups could be found and the loops can be sorted correctly. From our analysis, several conclusions regarding the optimal placement of anchor groups become obvious: (1) The correct placement of anchor groups is even more important for longer gap lengths, (2) medium length fragments (length 5-8) perform better than short or long ones, (3) the placement of anchor groups at hydrophobic amino acids gives a higher chance to include the best possible loop, (4) anchor groups within secondary structure elements, in particular beta-sheets are suitable, (5) amino acids with lower solvent accessibility are better anchor group. A preliminary test using a combination of the anchor group positioning criteria deduced from our analysis shows very promising results.     

A novel method for predicting and using distance constraints of high accuracy for refining protein structure prediction

The principal bottleneck in protein structure prediction is the refinement of models from lower accuracies to the resolution observed by experiment. We developed a novel constraints‐based refinement method that identifies a high number of accurate input constraints from initial models and rebuilds them using restrained torsion angle dynamics (rTAD). We previously created a Bayesian statistics‐based residue‐specific all‐atom probability discriminatory function (RAPDF) to discriminate native‐like models by measuring the probability of accuracy for atom type distances within a given model. Here, we exploit RAPDF to score (i.e., filter) constraints from initial predictions that may or may not be close to a native‐like state, obtain consensus of top scoring constraints amongst five initial models, and compile sets with no redundant residue pair constraints. We find that this method consistently produces a large and highly accurate set of distance constraints from which to build refinement models. We further optimize the balance between accuracy and coverage of constraints by producing multiple structure sets using different constraint distance cutoffs, and note that the cutoff governs spatially near versus distant effects in model generation. This complete procedure of deriving distance constraints for rTAD simulations improves the quality of initial predictions significantly in all cases evaluated by us. Our procedure represents a significant step in solving the protein structure prediction and refinement problem, by enabling the use of consensus constraints, RAPDF, and rTAD for protein structure modeling and refinement. Proteins 2009. © 2009 Wiley‐Liss, Inc.     

Reduced representation model of protein structure prediction: statistical potential and genetic algorithms.

A reduced representation model, which has been described in previous reports, was used to predict the folded structures of proteins from their primary sequences and random starting conformations. The molecular structure of each protein has been reduced to its backbone atoms (with ideal fixed bond lengths and valence angles) and each side chain approximated by a single virtual united-atom. The coordinate variables were the backbone dihedral angles phi and psi. A statistical potential function, which included local and nonlocal interactions and was computed from known protein structures, was used in the structure minimization. A novel approach, employing the concepts of genetic algorithms, has been developed to simultaneously optimize a population of conformations. With the information of primary sequence and the radius of gyration of the crystal structure only, and starting from randomly generated initial conformations, I have been able to fold melittin, a protein of 26 residues, with high computational convergence. The computed structures have a root mean square error of 1.66 A (distance matrix error = 0.99 A) on average to the crystal structure. Similar results for avian pancreatic polypeptide inhibitor, a protein of 36 residues, are obtained. Application of the method to apamin, an 18-residue polypeptide with two disulfide bonds, shows that it folds apamin to native-like conformations with the correct disulfide bonds formed.