Secondary Structure Class Research Articles

Ensemble of Neural Networks to Solve Class Imbalance Problem of Protein Secondary Structure Prediction

Protein secondary structures prediction (PSSP) is considered as a challenging task in bioinformatics. Many approaches have been proposed in last few decades in order to solve this problem. Despite the enhancements achieved, the prediction accuracy still remains limited. Accurate prediction of the secondary structure of proteins is a critical step in deducing tertiary structure of proteins and their functions. Among the proposed approaches to tackle this problem, artificial neural networks (ANN) are considered as one of the most successful methods that widely used in the field of PSSP. Recently, many efforts have been devoted to modify, improve and combine this technology with other machine learning methods in order to get better results. In this paper, we have proposed an ensemble method which combines the outputs of four feedforward neural networks. In each network one of the machine learning approaches has been applied in order to solve the class imbalance problem of protein secondary structure classification. The experimental results on RS126 data set show that our ensemble system has better performance compare to the best individual classifier. The results also reveal that the proposed system yields significant improvement in prediction accuracy of beta-sheet structure and a more balanced classification of three secondary structures.

Advances in the Prediction of Turn Structures in Peptides and Proteins.

Turns are essential for protein structure as they allow the polypeptide chain to fold backup on itself. They also occur within protein binding sites, at proteinprotein interfaces and in small bioactive peptides, where they can play a crucial role for molecular recognition. Turn structures are an important class of protein secondary structure, although relatively little attention is paid to them with respect to helices and β-sheets. Protein structure prediction, functional analysis of proteins and peptides, and computer-aided drug design could all benefit from making use of accurately predicted turn structures from amino acid sequence. Here, recent advances of turn structure prediction and the underlying turn classification will be discussed together with their applications.

Specific Residues of a Conserved Domain in the N Terminus of the Human Cytomegalovirus pUL50 Protein Determine Its Intranuclear Interaction with pUL53

Herpesviral capsids are assembled in the host cell nucleus and are subsequently translocated to the cytoplasm. During this process it has been demonstrated that the human cytomegalovirus proteins pUL50 and pUL53 interact and form, together with other viral and cellular proteins, the nuclear egress complex at the nuclear envelope. In this study we provide evidence that specific residues of a conserved N-terminal region of pUL50 determine its intranuclear interaction with pUL53. In silico evaluation and biophysical analyses suggested that the conserved region forms a regular secondary structure adopting a globular fold. Importantly, site-directed replacement of individual amino acids by alanine indicated a strong functional influence of specific residues inside this globular domain. In particular, mutation of the widely conserved residues Glu-56 or Tyr-57 led to a loss of interaction with pUL53. Consistent with the loss of binding properties, mutants E56A and Y57A showed a defective function in the recruitment of pUL53 to the nuclear envelope in expression plasmid-transfected and human cytomegalovirus-infected cells. In addition, in silico analysis suggested that residues 3-20 form an amphipathic α-helix that appears to be conserved among Herpesviridae. Point mutants revealed a structural role of this N-terminal α-helix for pUL50 stability rather than a direct role in the binding of pUL53. In contrast, the central part of the globular domain including Glu-56 and Tyr-57 is directly responsible for the functional interaction with pUL53 and thus determines formation of the basic nuclear egress complex.

Protein secondary structure prediction using DWKF based on SVR-NSGAII

Prediction of protein secondary structure is an important step towards elucidating its three dimensional structure and its function. This is a challenging problem in bioinformatics. By introduction of machine learning for protein structure prediction, a solution has brought to this challenge to some extent. In the literature of Machine learning or data mining, regression and classification problems are typically viewed as two distinct problems differentiated by continuous or categorical dependent variable. There are endeavors to use regression methods to solve the classification problem and vice versa. To regard a classification problem as a regression one, we proposed a method which is based on Support Vector Regression (SVR) classification model as one of the powerful methods in the field of machine intelligence. We applied non-dominated Sorting Genetic Algorithm II (NSGAII) to find mapping points (MPs) for rounding a real-value to an integer one. Also NSGAII is used for finding out and tuning SVR kernel parameters optimally to enhance the performance of our model and achieve better results. At the other hand, using a suitable SVR kernel function for a particular problem can improve the prediction results remarkably but there is not a kernel which can predict all protein secondary structure classes with acceptable accuracy. Therefore we use a Dynamic Weighted Kernel Fusion (DWKF) method for fusing of three SVR kernels to achieve a supreme performance. Also to improve our method, Position Scoring Matrix (PSSM) profiles are used as the input information to it. The goals of this research are to regulate SVR parameters and fuse different SVR kernel outputs in order to determine protein secondary structure classes accurately. The obtained classification accuracies of our method are 85.79% and 84.94% on RS126 and CB513 datasets respectively and they are promising with regard to other classification methods in the literature. Moreover, for gauging our method behavior in comparison to other state of arts methods, an independent dataset is used and achieves 81.4% accuracy. Our method cannot achieve the best value for any considered performance metrics on an independent dataset but its values for whole metrics are quite acceptable.

Protein Secondary Structure Prediction with SPARROW

A first step toward predicting the structure of a protein is to determine its secondary structure. The secondary structure information is generally used as starting point to solve protein crystal structures. In the present study, a machine learning approach based on a complete set of two-class scoring functions was used. Such functions discriminate between two specific structural classes or between a single specific class and the rest. The approach uses a hierarchical scheme of scoring functions and a neural network. The parameters are determined by optimizing the recall of learning data. Quality control is performed by predicting separate independent test data. A first set of scoring functions is trained to correlate the secondary structures of residues with profiles of sequence windows of width 15, centered at these residues. The sequence profiles are obtained by multiple sequence alignment with PSI-BLAST. A second set of scoring functions is trained to correlate the secondary structures of the center residues with the secondary structures of all other residues in the sequence windows used in the first step. Finally, a neural network is trained using the results from the second set of scoring functions as input to make a decision on the secondary structure class of the residue in the center of the sequence window. Here, we consider the three-class problem of helix, strand, and other secondary structures. The corresponding prediction scheme "SPARROW" was trained with the ASTRAL40 database, which contains protein domain structures with less than 40% sequence identity. The secondary structures were determined with DSSP. In a loose assignment, the helix class contains all DSSP helix types (α, 3-10, π), the strand class contains β-strand and β-bridge, and the third class contains the other structures. In a tight assignment, the helix and strand classes contain only α-helix and β-strand classes, respectively. A 10-fold cross validation showed less than 0.8% deviation in the fraction of correct structure assignments between true prediction and recall of data used for training. Using sequences of 140,000 residues as a test data set, 80.46% ± 0.35% of secondary structures are predicted correctly in the loose assignment, a prediction performance, which is very close to the best results in the field. Most applications are done with the loose assignment. However, the tight assignment yields 2.25% better prediction performance. With each individual prediction, we also provide a confidence measure providing the probability that the prediction is correct. The SPARROW software can be used and downloaded on the Web page http://agknapp.chemie.fu-berlin.de/sparrow/ .

Analysis of human collagen sequences.

The extracellular matrix is fast emerging as important component mediating cell-cell interactions, along with its established role as a scaffold for cell support. Collagen, being the principal component of extracellular matrix, has been implicated in a number of pathological conditions. However, collagens are complex protein structures belonging to a large family consisting of 28 members in humans; hence, there exists a lack of in depth information about their structural features. Annotating and appreciating the functions of these proteins is possible with the help of the numerous biocomputational tools that are currently available. This study reports a comparative analysis and characterization of the alpha-1 chain of human collagen sequences. Physico-chemical, secondary structural, functional and phylogenetic classification was carried out, based on which, collagens 12, 14 and 20, which belong to the FACIT collagen family, have been identified as potential players in diseased conditions, owing to certain atypical properties such as very high aliphatic index, low percentage of glycine and proline residues and their proximity in evolutionary history. These collagen molecules might be important candidates to be investigated further for their role in skeletal disorders.

PSSP with dynamic weighted kernel fusion based on SVM-PHGS

Since 1960s, researchers have proposed several prediction methods, for protein secondary structure prediction (PSSP), whereas the accuracy of them is no more than 80%. In this case, there is an urgent need to introduce a high accuracy prediction method. One learning method called support vector machines (SVMs) has shown comparable or better results than neural networks on bioinformatics applications. This research proposes a method based on SVM which has been improved by a new parallel multi class (PMC) method, parallel hierarchical grid search (PHGS), cross validation (CV) technique and weighted kernel fusion (WKF) method. The presented PHGS has been applied to regularize parameters of SVM’s kernel function which have an important impact on the accuracy. Using a suitable input data and kernel function for a particular problem can improve the prediction results remarkably. Also to improve our method, Position Scoring Matrix (PSSM) profiles are used as the input information to it. The goals of this study are to calibrate kernel function parameters and fusion of different kernel functions’ result in order to determine protein secondary structure classes accurately. The right choice of a fusion method is an important issue in creating a supreme performance so we propose a dynamic weight allocation method based on a non-linear analysis system. The obtained classification accuracies of our method are 84.65% and 83.94% on RS126 and CB513 datasets respectively and they are very promising with regard to other classification methods in the literature for this problem. Also for evaluating our method behavior in comparison to other state of arts methods, an independent dataset is used. The results show that the comprehensibility of WKF based on SVM-PHGS is much better than other methods.

Random generation of RNA secondary structures according to native distributions.

BackgroundRandom biological sequences are a topic of great interest in genome analysis since, according to a powerful paradigm, they represent the background noise from which the actual biological information must differentiate. Accordingly, the generation of random sequences has been investigated for a long time. Similarly, random object of a more complicated structure like RNA molecules or proteins are of interest.ResultsIn this article, we present a new general framework for deriving algorithms for the non-uniform random generation of combinatorial objects according to the encoding and probability distribution implied by a stochastic context-free grammar. Briefly, the framework extends on the well-known recursive method for (uniform) random generation and uses the popular framework of admissible specifications of combinatorial classes, introducing weighted combinatorial classes to allow for the non-uniform generation by means of unranking. This framework is used to derive an algorithm for the generation of RNA secondary structures of a given fixed size. We address the random generation of these structures according to a realistic distribution obtained from real-life data by using a very detailed context-free grammar (that models the class of RNA secondary structures by distinguishing between all known motifs in RNA structure). Compared to well-known sampling approaches used in several structure prediction tools (such as SFold) ours has two major advantages: Firstly, after a preprocessing step in time for the computation of all weighted class sizes needed, with our approach a set of m random secondary structures of a given structure size n can be computed in worst-case time complexity while other algorithms typically have a runtime in . Secondly, our approach works with integer arithmetic only which is faster and saves us from all the discomforting details of using floating point arithmetic with logarithmized probabilities.ConclusionA number of experimental results shows that our random generation method produces realistic output, at least with respect to the appearance of the different structural motifs. The algorithm is available as a webservice at http://wwwagak.cs.uni-kl.de/NonUniRandGen and can be used for generating random secondary structures of any specified RNA type. A link to download an implementation of our method (in Wolfram Mathematica) can be found there, too.

Effect of Neural Network Parameters on RNA Secondary Structure Classification

Helix, Hairpin, Bulge, external loop, internal loop, multi-branch loop are the elements of RNA secondary structure. We have designed a neural network to classify the RNA sequence in to three categories i.e Hairpin, helix, neither of two. This can be extended to classify into all secondary structure elements. If all the elements are predicted then we can determine the entire structure of a RNA family. The parameters of neural network affect the performance of the network. But there are no rules to define the value of these parameters of network. For a given problem the optimal value of parameters can be obtained by performing the experiments on their values. This paper shows the effect on the performance of classification by varying the number of hidden layers, number of neurons and activation functions.

Learning sparse models for a dynamic Bayesian network classifier of protein secondary structure.

BackgroundProtein secondary structure prediction provides insight into protein function and is a valuable preliminary step for predicting the 3D structure of a protein. Dynamic Bayesian networks (DBNs) and support vector machines (SVMs) have been shown to provide state-of-the-art performance in secondary structure prediction. As the size of the protein database grows, it becomes feasible to use a richer model in an effort to capture subtle correlations among the amino acids and the predicted labels. In this context, it is beneficial to derive sparse models that discourage over-fitting and provide biological insight.ResultsIn this paper, we first show that we are able to obtain accurate secondary structure predictions. Our per-residue accuracy on a well established and difficult benchmark (CB513) is 80.3%, which is comparable to the state-of-the-art evaluated on this dataset. We then introduce an algorithm for sparsifying the parameters of a DBN. Using this algorithm, we can automatically remove up to 70-95% of the parameters of a DBN while maintaining the same level of predictive accuracy on the SD576 set. At 90% sparsity, we are able to compute predictions three times faster than a fully dense model evaluated on the SD576 set. We also demonstrate, using simulated data, that the algorithm is able to recover true sparse structures with high accuracy, and using real data, that the sparse model identifies known correlation structure (local and non-local) related to different classes of secondary structure elements.ConclusionsWe present a secondary structure prediction method that employs dynamic Bayesian networks and support vector machines. We also introduce an algorithm for sparsifying the parameters of the dynamic Bayesian network. The sparsification approach yields a significant speed-up in generating predictions, and we demonstrate that the amino acid correlations identified by the algorithm correspond to several known features of protein secondary structure. Datasets and source code used in this study are available at http://noble.gs.washington.edu/proj/pssp.

Rosetta FlexPepDock ab-initio: simultaneous folding, docking and refinement of peptides onto their receptors.

Flexible peptides that fold upon binding to another protein molecule mediate a large number of regulatory interactions in the living cell and may provide highly specific recognition modules. We present Rosetta FlexPepDock ab-initio, a protocol for simultaneous docking and de-novo folding of peptides, starting from an approximate specification of the peptide binding site. Using the Rosetta fragments library and a coarse-grained structural representation of the peptide and the receptor, FlexPepDock ab-initio samples efficiently and simultaneously the space of possible peptide backbone conformations and rigid-body orientations over the receptor surface of a given binding site. The subsequent all-atom refinement of the coarse-grained models includes full side-chain modeling of both the receptor and the peptide, resulting in high-resolution models in which key side-chain interactions are recapitulated. The protocol was applied to a benchmark in which peptides were modeled over receptors in either their bound backbone conformations or in their free, unbound form. Near-native peptide conformations were identified in 18/26 of the bound cases and 7/14 of the unbound cases. The protocol performs well on peptides from various classes of secondary structures, including coiled peptides with unusual turns and kinks. The results presented here significantly extend the scope of state-of-the-art methods for high-resolution peptide modeling, which can now be applied to a wide variety of peptide-protein interactions where no prior information about the peptide backbone conformation is available, enabling detailed structure-based studies and manipulation of those interactions.

Aptamer-Mediated Inhibition of Mycobacterium tuberculosis Polyphosphate Kinase 2

Inorganic polyphosphate (polyP) plays a number of critical roles in bacterial persistence, stress, and virulence. PolyP intracellular metabolism is regulated by the polyphosphate kinase (PPK) protein families, and inhibition of PPK activity is a potential approach to disrupting polyP-dependent processes in pathogenic organisms. Here, we biochemically characterized Mycobacterium tuberculosis (MTB) PPK2 and developed DNA-based aptamers that inhibit the enzyme's catalytic activities. MTB PPK2 catalyzed polyP-dependent phosphorylation of ADP to ATP at a rate 838 times higher than the rate of polyP synthesis. Gel filtration chromatography suggested MTB PPK2 to be an octamer. DNA aptamers were isolated against MTB PPK2. Circular dichroism revealed that aptamers grouped into two distinct classes of secondary structure; G-quadruplex and non-G-quadruplex. A selected G-quadruplex aptamer was highly selective for binding to MTB PPK2 with a dissociation constant of 870 nM as determined by isothermal titration calorimetry. The binding between MTB PPK2 and the aptamer was exothermic yet primarily driven by entropy. This G-quadruplex aptamer inhibited MTB PPK2 with an IC(50) of 40 nM and exhibited noncompetitive inhibition kinetics. Mutational mechanistic analysis revealed an aptamer G-quadruplex motif is critical for enzyme inhibition. The aptamer was also tested against Vibrio cholerae PPK2, where it showed an IC(50) of 105 nM and insignificant inhibition against more distantly related Laribacter hongkongensis PPK2.

Application of ring‐closing metathesis to Grb2 SH3 domain‐binding peptides

Molecular processes depending on protein–protein interactions can use consensus recognition sequences that possess defined secondary structures. Left-handed polyproline II (PPII) helices are a class of secondary structure commonly involved with cellular signal transduction. However, unlike -helices, for which a substantial body of work exists regarding applications of ring-closing metathesis (RCM), there are few reports on the stabilization of PPII helices by RCM methodologies. The current study examined the effects of RCM macrocyclization on left-handed PPII helices involved with the SH3 domain-mediated binding of Sos1–Grb2. Starting with the Sos1-derived peptide “Ac-V1-P2-P3-P4-V5-P6-P7-R8-R9-R10-amide,” RCM macrocyclizations were conducted using alkenyl chains of varying lengths originating from the pyrrolidine rings of the Pro4 and Pro7 residues. The resulting macrocyclic peptides showed increased helicity as indicated by circular dichroism and enhanced abilities to block Grb2–Sos1 interactions in cell lysate pull-down assays. The synthetic approach may be useful in RCM macrocyclizations, where maintenance of proline integrity at both ring junctures is desired.

K-Partite RNA Secondary Structures

RNA secondary structure prediction is a fundamental problem in structural bioinformatics. The prediction problem is difficult because RNA secondary structures may contain pseudoknots formed by crossing base pairs. We introduce k-partite secondary structures as a simple classification of RNA secondary structures with pseudoknots. An RNA secondary structure is k-partite if it is the union of k pseudoknot-free sub-structures. Most known RNA secondary structures are either bipartite or tripartite. We show that there exists a constant number k such that any secondary structure can be modified into a k-partite secondary structure with approximately the same free energy. This offers a partial explanation of the prevalence of k-partite secondary structures with small k. We give a complete characterization of the computational complexities of recognizing k-partite secondary structures for all k > or = 2, and show that this recognition problem is essentially the same as the k-colorability problem on circle graphs. We present two simple heuristics, iterated peeling and first-fit packing, for finding k-partite RNA secondary structures. For maximizing the number of base pair stackings, our iterated peeling heuristic achieves a constant approximation ratio of at most k for 2 < or = k < or = 5, and at most [Formula: see text] for k > or = 6. Experiment on sequences from PseudoBase shows that our first-fit packing heuristic outperforms the leading method HotKnots in predicting RNA secondary structures with pseudoknots. Supplementary Material can be found at www.libertonline.com.

ASYMPTOTICS OF CANONICAL AND SATURATED RNA SECONDARY STRUCTURES

It is a classical result of Stein and Waterman that the asymptotic number of RNA secondary structures is 1.104366 · n-3/2 · 2.618034n. In this paper, we study combinatorial asymptotics for two special subclasses of RNA secondary structures — canonical and saturated structures. Canonical secondary structures are defined to have no lonely (isolated) base pairs. This class of secondary structures was introduced by Bompfünewerer et al., who noted that the run time of Vienna RNA Package is substantially reduced when restricting computations to canonical structures. Here we provide an explanation for the speed-up, by proving that the asymptotic number of canonical RNA secondary structures is 2.1614 · n-3/2 · 1.96798n and that the expected number of base pairs in a canonical secondary structure is 0.31724 · n. The asymptotic number of canonical secondary structures was obtained much earlier by Hofacker, Schuster and Stadler using a different method. Saturated secondary structures have the property that no base pairs can be added without violating the definition of secondary structure (i.e. introducing a pseudoknot or base triple). Here we show that the asymptotic number of saturated structures is 1.07427 · n-3/2 · 2.35467n, the asymptotic expected number of base pairs is 0.337361 · n, and the asymptotic number of saturated stem-loop structures is 0.323954 · 1.69562n, in contrast to the number 2n - 2 of (arbitrary) stem-loop structures as classically computed by Stein and Waterman. Finally, we apply the work of Drmota to show that the density of states for [all resp. canonical resp. saturated] secondary structures is asymptotically Gaussian. We introduce a stochastic greedy method to sample random saturated structures, called quasi-random saturated structures, and show that the expected number of base pairs is 0.340633 · n.

On quantitative effects of RNA shape abstraction

Over the last few decades, much effort has been taken to develop approaches for identifying good predictions of RNA secondary structure. This is due to the fact that most computational prediction methods based on free energy minimization compute a number of suboptimal foldings and we have to identify the native folding among all these possible secondary structures. Using the abstract shapes approach as introduced by Giegerich et al. (Nucleic Acids Res 32(16):4843-4851, 2004), each class of similar secondary structures is represented by one shape and the native structures can be found among the top shape representatives. In this article, we derive some interesting results answering enumeration problems for abstract shapes and secondary structures of RNA. We compute precise asymptotics for the number of different shape representations of size n and for the number of different shapes showing up when abstracting from secondary structures of size n under a combinatorial point of view. A more realistic model taking primary structures into account remains an open challenge. We give some arguments why the present techniques cannot be applied in this case.

Flow Linear Dichroism of Some Prototypical Proteins

Flow linear dichroism (LD) spectroscopy provides information on the orientation of molecules in solution and hence on the relative orientation of parts of molecules. Long molecules such as fibrous proteins can be aligned in Couette flow cells and characterized using LD. We have measured using Couette flow and calculated from first principles the LD of proteins representing prototypical secondary structure classes: a self-assembling fiber and tropomyosin (all-alpha-helical), FtsZ (an alphabeta protein), an amyloid fibril (beta-sheet), and collagen [poly(proline)II helices]. The combination of calculation and experiment allows elucidation of the protein orientation in the Couette flow and the orientation of chromophores within the protein fibers.

Analysis of the secondary structure of a protein's N-terminal

The major protein component from aleurone cells of barley (Hordeum vulgare L.), PACB, is related to 7S globulins present in other cereals and to the vicilin-type 7S globulins of legumes and cotton seed. It contains 4 subunits of about 20, 25, 40 and 50 kDa molecular weights. The N-terminal sequence of 16 amino acids (over 260 atoms) in the protein was previously determined, and our aim is the prediction of its secondary structure. The empirical Chou-Fasman method was applied in an improved version as well as the empirical DSC method (discrimination of protein secondary structure class) with quite similar results. A molecular dynamics simulation was also performed, using the FF99SB forcefield within AMBER version 9.0. Solvation effects were incorporated using the Born model. The results are compared and a 3D model is proposed.

Differential Inhibitory Activities and Stabilisation of DNA Aptamers against the SARS Coronavirus Helicase

The helicase from severe acute respiratory syndrome coronavirus (SARS‐CoV) possesses NTPase, duplex RNA/DNA‐unwinding and RNA‐capping activities that are essential for viral replication and proliferation. Here, we have isolated DNA aptamers against the SARS‐CoV helicase from a combinatorial DNA library. These aptamers show two distinct classes of secondary structure, G‐quadruplex and non‐G‐quadruplex, as shown by circular dichroism and gel electrophoresis. All of the aptamers that were selected stimulated ATPase activity of the SARS‐CoV helicase with low‐nanomolar apparent K m values. Intriguingly, only the non‐G‐quadruplex aptamers showed specific inhibition of helicase activities, whereas the G‐quadruplex aptamers did not inhibit helicase activities. The non‐G‐quadruplex aptamer with the strongest inhibitory potency was modified at the 3′‐end with biotin or inverted thymidine, and the modification increased its stability in serum, particularly for the inverted thymidine modification. Structural diversity in selection coupled to post‐selection stabilisation has provided new insights into the aptamers that were selected for a helicase target. These aptamers are being further developed to inhibit SARS‐CoV replication.

Protein secondary structure class assignment on the basis of a new graphic representation

AbstractA novel 2D representation (M‐curve) has been provided to visualize the structure information of protein secondary structure sequences: (1) end point of the curve reflects difference of amino acid residue numbers in α‐helices and β‐strands of a protein; (2) Up/down ladder‐like structures in the curve show that α‐helices/β‐strands appear sequentially in this region; (3) triangular‐like/trapeziform‐like structures of the curve show that α‐helices and β‐strands appear alternatively in this region. So from the M‐curve, a protein could be directly assigned into corresponding secondary structure class. Moreover, a new numerical descriptor, four‐component vector, is introduced and applied to cluster analysis. © 2008 Wiley Periodicals, Inc. Int J Quantum Chem, 2009