Simultaneous prediction of protein secondary structure and transmembrane spans

Secondary structure of proteins refers to local and repetitive conformations, such as α-helices and β-strands, which occur in protein structures. Computational prediction of secondary structure from protein sequences has a long history with three generations of predictive methods. This unit summarizes several recent third-generation predictors. We discuss their inputs and outputs, availability, and predictive performance and explain how to perform and interpret their predictions. We cover methods for the prediction of the 3-class secondary structure states (helix, strand, and coil) as well as the 8-class secondary structure states. Recent empirical assessments and our small-scale analysis reveal that these predictions are characterized by high levels of accuracy, between 70% and 80%. We emphasize that modern predictors are available to end users in the form of convenient-to-use Web servers and stand-alone software. © 2016 by John Wiley & Sons, Inc.

The predictive limits of the amino acid composition for the secondary structural content (percentage of residues in the secondary structural states helix, sheet, and coil) in proteins are assessed quantitatively. For the first time, techniques for prediction of secondary structural content are presented which rely on the amino acid composition as the only information on the query protein. In our first method, the amino acid composition of an unknown protein is represented by the best (in a least square sense) linear combination of the characteristic amino acid compositions of the three secondary structural types computed from a learning set of tertiary structures. The second technique is a generalization of the first one and takes into account also possible compositional couplings between any two sorts of amino acids. Its mathematical formulation results in an eigenvalue/eigenvector problem of the second moment matrix describing the amino acid compositional fluctuations of secondary structural types in various proteins of a learning set. Possible correlations of the principal directions of the eigenspaces with physical properties of the amino acids were also checked. For example, the first two eigenvectors of the helical eigenspace correlate with the size and hydrophobicity of the residue types respectively. As learning and test sets of tertiary structures, we utilized representative, automatically generated subsets of Protein Data Bank (PDB) consisting of non-homologous protein structures at the resolution thresholds < or = 1.8A, < or = 2.0A, < or = 2.5A, and < or = 3.0 A. We show that the consideration of compositional couplings improves prediction accuracy, albeit not dramatically. Whereas in the self-consistency test (learning with the protein to be predicted), a clear decrease of prediction accuracy with worsening resolution is observed, the jackknife test (leave the predicted protein out) yielded best results for the largest dataset (< or = 3.0A, almost no difference to the self-consistency test!), i.e., only this set, with more than 400 proteins, is sufficient for stable computation of the parameters in the prediction function of the second method. The average absolute error in predicting the fraction of helix, sheet, and coil from amino acid composition of the query protein are 13.7, 12.6, and 11.4%, respectively with r.m.s. deviations in the range of 8.6 divided by 11.8% for the 3.0 A dataset in a jackknife test. The absolute precision of the average absolute errors is in the range of 1 divided by 3% as measured for other representative subsets of the PDB. Secondary structural content prediction methods found in the literature have been clustered in accordance with their prediction accuracies. To our surprise, much more complex secondary structure prediction methods utilized for the same purpose of secondary structural content prediction achieve prediction accuracies very similar to those of the present analytic techniques, implying that all the information beyond the amino acid composition is, in fact, mainly utilized for positioning the secondary structural state in the sequence but not for determination of the overall number of residues in a secondary structural type. This result implies that higher prediction accuracies cannot be achieved relying solely on the amino acid composition of an unknown query protein as prediction input. Our prediction program SSCP has been made available as a World Wide Web and E-mail service.

Combining Prediction of Protein Aggregation Propensities with Prediction of Other One-Dimensional Properties

Background: The Mycobacterium tuberculosis complex (MTBC) bacteria include the slowly growing, host-associated bacteria Mycobacterium tuberculosis, Mycobacterium Bovis, Mycobacterium microti, Mycobacterium africanum, Mycobacterium pinnipedii. Aim: Comparative Functional Genomics Studies for understanding the Hypothetical Proteins in Mycobacterium tuberculosis variant microti 12. Objective: A computational genomics study was performed to understand the 247 hypothetical protein genes. Functional annotation of virtual proteins was performed on different servers to maximize confidence level. Methods: Sequence Retrieval. The whole genome sequences for the Mycobacterium tuberculosis micro variant 12 were retrieved from the KEGG database ( http://www.genome.jp/kegg/) and were used for screening 247 hypothetical proteins (Fig. 1 ). Functional Annotation and Sub-cellular localization. The Mycobacterium tuberculosis micro variant 12 hypothetical proteins were screened and sorted out from the genome and were individually analyzed for the presence of conserved functional domains by using computational biology tools like CDD-BLAST ( https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) ;Pfam ( http://pfam.xfam.org/ncbiseq/398365647); The subcellular localization of hypothetical proteins was determined by CELLO2GO ( http://cello.life.nctu.edu.tw). These web tools can search the defined conserved domains in the sequences available in the online servers or databases and assist in the classification of proteins in the appropriate families. Protein Structure Prediction. The in-silico structure predictions of the hypothetical protein sequences showing functional properties were carried out by using the PS2 Protein Structure Prediction Server ( http://www.ps2.life.nctu.edu.tw/). The online server helps to generate the 3D structures of the hypothetical proteins. The server accepts the sequences in FASTA format as a query to generate resultant proteins 3D structures. The structure determination is completely based on the conserved template regions detected during functional annotations. Protein-protein interaction through String database: The interaction of each hypothetical protein analyzed for functional characteristics was subjected to a protein-protein interaction server for the prediction of a possible functional role in interaction amongst the available known proteins ( https://string-db.org/). This information can help us to further validated the functional role of such hypothetical proteins and their possible role in the Mycobacterium Tuberculosis micro variant. Protein secondary structure prediction through JPred4: The secondary structure prediction of all the hypothetical proteins was determined through JPred4 ( http://www.compbio.dundee.ac.uk/jpred4/index.html) and served to identify the available secondary structures in the unknown hypothetical protein sequences. These further help us to understand the available templates in the uncharacterized protein sequences for the prediction of novel functions associated with these proteins. The predictions were further characterized by the Phyre2 server for structural modeling and prediction of templates based on comparative analysis based on conserved domains. Protein modeling, prediction, and analysis through Phyre2. The hypothetical proteins which were identified to have functional properties were further characterized by the Phyre2 server ( http://www.sbg.bio.ic.ac.uk/phyre2) for structural modeling and prediction of templates based on comparative analysis based on conserved domains. Results: A computational genomics study was performed to understand the 247 hypothetical protein genes Functional annotation of virtual proteins, and was performed on different servers to maximize confidence level. The functional prediction was performed by CDD-Blast and Pfam. The gene sequences of proteins have probably been successfully functionally annotated, characterized, and their subcellular localization and 3-D structural predictions have been predicted computationally. Online automated bioinformatics tools such as CDD-Blast, Pfam, CELLO2GO and PS2-Server were used for the structural and functional characterization of screened hypothetical proteins. The structure, function, and subcellular localization of a hypothetical protein from Mycobacterium tuberculosis variant microti 12 have been obtained and presented (Fig. 2 ). Also, the three-dimensional structure generated after using the template with the highest score was displayed as the template ID in the structure column of the respective hypothetical protein. However, as systems biology denies hypothetical protein functions, the structures of such proteins can be tested through biological processes and experiments, making them suitable for understanding their role in the life cycle, pathogenesis, and drug development. We can further explore these predictive possibilities in pharmaceuticals, and other clinically relevant studies. This study by HP helped find structure-function relationships in Mycobacterium tuberculosis variant microti 12 using a variety of bioinformatics tools. The string database made predictions about protein-protein interactions and the template helped us predict a hypothetical protein structure and even helped us find its 3D protein structure. Protein profiling can be performed on structures retrieved from these servers. This is useful for proteomics studies, including protein-protein interactions, protein expression of specific hypothetical proteins, and post-translational modifications of protein-coding genes. Further understanding of these hypothetical proteins can help us to know more about the Mycobacterium tuberculosis complex (MTBC) and may assist in Drugs and inhibitors against different pathogens within this complex. Conclusion: The all-inclusive bioinformatic study has helped to functionally elucidate 247 hypothetical proteins, which have resulted and made it easier to understand many functional proteins available in the Mycobacterium tuberculosis micro variant 12. The subcellular localization of the 247 sorted hypothetical proteins was also carried &amp;amp; which further helped us understand the localization of identified enzymes or proteins. We have successfully characterized the 247 unknown proteins of hypothetical protein sequences from Mycobacterium tuberculosis micro variant 12 to validate their structure and functions of the gene products. These predicted functions and three-dimensional structures may lead to establishing their role in the life cycle of the bacterium. This computationally generated data can also be further used for developing new protocols for new vaccines against Mycobacterium tuberculosis micro variant 12 that are essential for preventing infection, diseases, and transmission. This complete result of Hypothetical Protein is needed for further studies of the whole genomic of the Mycobacterium Tuberculosis micro variant 12 for their function interpretation which further help in the understanding of its functions as well as structure. Moreover, this interpretation would help us to study the evolution of Mycobacterium Tuberculosis micro variant 12 which further helps in the process of discovering the drugs to inhibit the causes of diseases.

Protein secondary structure predictions and amino acid long range contact map predictions from primary sequence of proteins have been explored to aid in modelling protein tertiary structures. In order to evaluate the usefulness of secondary structure and 3D-residue contact prediction methods to model protein structures we have used the known Q3 (alpha-helix,beta-strands and irregular turns/loops) secondary structure information, along with residue-residue contact information as restraints for MODELLER. We present here results of our modelling studies on 30 best resolved single domain protein structures of varied lengths. The results shows that it is very difficult to obtain useful models even with 100% accurate secondary structure predictions and accurate residue contact predictions for up to 30% of residues in a sequence. The best models that we obtained for proteins of lengths 37, 70, 118, 136 and 193 amino acid residues are of RMSDs 4.17, 5.27, 9.12, 7.89 and 9.69,respectively. The results show that one can obtain better models for the proteins which have high percent of alpha-helix content. This analysis further shows that MODELLER restrain optimization program can be useful only if we have truly homologous structure(s) as a template where it derives numerous restraints, almost identical to the templates used. This analysis also clearly indicates that even if we satisfy several true residue-residue contact distances, up to 30%of their sequence length with fully known secondary structural information, we end up predicting model structures much distant from their corresponding native structures.

Sequence-based prediction of one dimensional structural properties of proteins has been a long-standing subproblem of protein structure prediction. Recently, prediction accuracy has been significantly improved due to the rapid expansion of protein sequence and structure libraries and advances in deep learning techniques, such as residual convolutional networks (ResNets) and Long-Short-Term Memory Cells in Bidirectional Recurrent Neural Networks (LSTM-BRNNs). Here we leverage an ensemble of LSTM-BRNN and ResNet models, together with predicted residue-residue contact maps, to continue the push towards the attainable limit of prediction for 3- and 8-state secondary structure, backbone angles (θ, τ, ϕ and ψ), half-sphere exposure, contact numbers and solvent accessible surface area (ASA). The new method, named SPOT-1D, achieves similar, high performance on a large validation set and test set (≈1000 proteins in each set), suggesting robust performance for unseen data. For the large test set, it achieves 87% and 77% in 3- and 8-state secondary structure prediction and 0.82 and 0.86 in correlation coefficients between predicted and measured ASA and contact numbers, respectively. Comparison to current state-of-the-art techniques reveals substantial improvement in secondary structure and backbone angle prediction. In particular, 44% of 40-residue fragment structures constructed from predicted backbone Cα-based θ and τ angles are less than 6 Å root-mean-squared-distance from their native conformations, nearly 20% better than the next best. The method is expected to be useful for advancing protein structure and function prediction. SPOT-1D and its data is available at: http://sparks-lab.org/. Supplementary data are available at Bioinformatics online.

Prediction of protein secondary structure and active sites using the alignment of homologous sequences

Background: Protein secondary structure prediction is an important task in bioinformatics and structural biology. Protein’s structure is the basis for its corresponding function. Experimental methods for determining the tertiary structure of proteins are both costly and time-consuming. Since the tertiary structure of proteins is further formed by secondary structure, leveraging computational approaches for efficient prediction of protein secondary structure is important. Both local and global interactions between amino acids affect the prediction results. Objective: We propose a module aimed at processing sequence profile features for deep feature extraction and constructing a lightweight network to extract fused features. Methods: To enhance the network’s ability to capture both local and global interactions, we propose an efficient method InConTPSS, which integrates convolution operation with different receptive fields and temporal convolutional networks in the inception architecture. Concurrently, InConTPSS takes into account the issue of distribution imbalance across various states of secondary structures and improves the predictive performance of scarce categories. Results: Experimental results on six benchmark datasets (including CASP12, CASP13, CASP14, CB513, TEST2016, and TEST2018) demonstrate our method achieves state-of-the-art performance with a simpler model on both 3-state and 8-state secondary structure prediction. result: Experimental results on benchmark datasets (including CASP12, CASP13, CASP14, CB513, TEST2016 and TEST2018) demonstrate our method achieves state-of-the-art performance with a simpler model on both 3-state and 8-state secondary structure prediction. Conclusion: Through the combination of the convolutional layer and temporal convolutional network, the inception network structure can effectively process the fused features and improve the prediction results. InConTPSS achieves the most advanced performance in protein secondary structure prediction, and the reasonable use of label-distribution-aware margin loss in our method can effectively improve the prediction accuracy of scarce secondary structures.

Prediction of protein secondary structure provides information that is useful for other prediction methods like fold recognition and ab initio 3D prediction. A consensus prediction constructed from the output of several methods should yield more reliable results than each of the individual methods. We present an approach that reveals subtle but systematic differences in the output of different secondary structure prediction methods allowing the derivation of coherent consensus predictions. The method uses a machine learning technique that builds decision trees from existing data. The first results of our analysis show that consensus prediction of protein secondary structure may be improved both quantitatively and qualitatively.

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

Circular dichroism (CD) is a spectroscopic technique commonly used to investigate the structure of proteins. Major secondary structure types, alpha-helices and beta-strands, produce distinctive CD spectra. Thus, by comparing the CD spectrum of a protein of interest to a reference set consisting of CD spectra of proteins of known structure, predictive methods can estimate the secondary structure of the protein. Currently available methods, including K2D2, use such experimental CD reference sets, which are very small in size when compared to the number of tertiary structures available in the Protein Data Bank (PDB). Conversely, given a PDB structure, it is possible to predict a theoretical CD spectrum from it. The methodological framework for this calculation was established long ago but only recently a convenient implementation called DichroCalc has been developed. In this study, we set to determine whether theoretically derived spectra could be used as reference set for accurate CD based predictions of secondary structure. We used DichroCalc to calculate the theoretical CD spectra of a nonredundant set of structures representing most proteins in the PDB, and applied a straightforward approach for predicting protein secondary structure content using these theoretical CD spectra as reference set. We show that this method improves the predictions, particularly for the wavelength interval between 200 and 240 nm and for beta-strand content. We have implemented this method, called K2D3, in a publicly accessible web server at http://www. ogic.ca/projects/k2d3.

Secondary structure predictions are increasingly becoming the workhorse for several methods aiming at predicting protein structure and function. Here we use ensembles of bidirectional recurrent neural network architectures, PSI-BLAST-derived profiles, and a large nonredundant training set to derive two new predictors: (a) the second version of the SSpro program for secondary structure classification into three categories and (b) the first version of the SSpro8 program for secondary structure classification into the eight classes produced by the DSSP program. We describe the results of three different test sets on which SSpro achieved a sustained performance of about 78% correct prediction. We report confusion matrices, compare PSI-BLAST to BLAST-derived profiles, and assess the corresponding performance improvements. SSpro and SSpro8 are implemented as web servers, available together with other structural feature predictors at: http://promoter.ics.uci.edu/BRNN-PRED/.

The solution of binary classification problems using the Support Vector Machine (SVM) method has been well developed. Though multi-class classification is typically solved by combining several binary classifiers, recently, several multi-class methods that consider all classes at once have been proposed. However, these methods require resolving a much larger optimization problem and are applicable to small datasets. Three methods based on binary classifications: one-against-all (OAA), one-against-one (OAO), and directed acyclic graph (DAG), and two approaches for multi-class problem by solving one single optimization problem, are implemented to predict protein secondary structure. Our experiments indicate that multi-class SVM methods are more suitable for protein secondary structure (PSS) prediction than the other methods, including binary SVMs, because their capacity to solve an optimization problem in one step. Furthermore, in this paper, we argue that it is feasible to extend the prediction accuracy by adding a second-stage multi-class SVM to capture the contextual information among secondary structural elements and thereby further improving the accuracies. We demonstrate that two-stage SVMs perform better than single-stage SVM techniques for PSS prediction using two datasets and report a maximum accuracy of 79.5%.

In the field of bioinformatics, the prediction of secondary structure of proteins is very important. It can be obtained from the prediction of primary structure (amino acid sequence) and can provide reference for the prediction of tertiary structure of proteins. Amino acid sequences of proteins are encoded with several features and then combined into the prediction network. Convolutional neural network has excellent performance in text and sequence information extraction. The amino acid sequence of protein is also a special sequence, so the convolutional neural network can be used to extract the information in the sequence. Moreover, the influence of amino acids on the formation of secondary structure varies with different distances, so in the experiment, convolutional neural networks with convolution nuclei of different sizes were used to form multi-scale convolution blocks to extract amino acid sequence information. At the same time, the sliding window technique is also used to show the interaction between the sequences, and a long amino acid sequence is divided into some amino acid fragments and input into the model. Finally, the accuracy of Q8 on the dataset CB6133_filtered reaches 71%.

Secondary structure prediction is an invaluable tool in determining the 3-D structure and function of proteins. Typically, protein secondary structure prediction methods suffer from low accuracy in beta-strand predictions, where nonlocal interactions play a significant role. There is a considerable need to model such long- range interactions that contribute to the stabilization of a protein molecule. In this paper, we introduce an alternative decoding technique for the hidden semi-Markov model (HSMM) originally employed in the BSPSS algorithm, and further developed in the IPSSP algorithm. The proposed method is based on the N-best paradigm where a set of most likely segmentations is computed. To generate suboptimal segmentations (i.e., alternative prediction sequences), we developed two N-best search algorithms. The first one is an A* stack decoder algorithm that extends paths (or hypotheses) by one symbol at each iteration. The second algorithm locally keeps the end positions of the highest scoring K previous segments and performs backtracking. Both algorithms employ the hidden semi- Markov model described in Aydin etal. [5], and use Viterbi scoring to compute the N-best list. The availability of near-optimal segmentations and the utilization of the Viterbi scoring enable the sequences to be rescored using more complex dependency models that characterize nonlocal interactions in beta-sheets. After the score update, one can either keep the segmentations to be employed in 3-D structure prediction or predict the secondary structure by applying a weighted voting procedure to a set of top scoring M ges 1 segmentations. The accuracy measures of the N-best method when used to predict the secondary structure are shown to be comparable or better than the classical Viterbi decoder (MAP estimator), tested under the single-sequence condition. When no rescoring is applied, the stack decoder algorithm with sufficiently large M improves the overall sensitivity measure (Q3) of the Viterbi algorithm by 1.1%. At the same M value, the N-best Viterbi algorithm improves the Q3 measure by 0.25% as well as the sensitivity measures specific for each secondary structure type (Qobs alpha, Qobs beta, Qobs L). When the sequences are rescored using the posterior probability distribution computed by the posterior decoding algorithm (MPM estimator), N-best Viterbi improves the Q3 measure of the Viterbi algorithm by 2.6%. The rescored N-best list approach also enables us to generate suboptimal segmentations that are valid sequences (i.e., realizable from the hidden semi-Markov model). Although the N-best algorithms and the score update procedure brought significant improvements over the Viterbi algorithm, they were not able to outperform the posterior decoding algorithm in the single-sequence condition. Further improvements in the prediction accuracy should be possible with the incorporation of sophisticated models for nonlocal interactions and other physical constraints that stabilize the overall structure of a protein.