Pcons: a neural-network-based consensus predictor that improves fold recognition

During recent years many protein fold recognition methods have been developed, based on different algorithms and using various kinds of information. To examine the performance of these methods several evaluation experiments have been conducted. These include blind tests in CASP/CAFASP, large scale benchmarks, and long-term, continuous assessment with newly solved protein structures. These studies confirm the expectation that for different targets different methods produce the best predictions, and the final prediction accuracy could be improved if the available methods were combined in a perfect manner. In this article a neural-network-based consensus predictor, Pcons, is presented that attempts this task. Pcons attempts to select the best model out of those produced by six prediction servers, each using different methods. Pcons translates the confidence scores reported by each server into uniformly scaled values corresponding to the expected accuracy of each model. The translated scores as well as the similarity between models produced by different servers is used in the final selection. According to the analysis based on two unrelated sets of newly solved proteins, Pcons outperforms any single server by generating approximately 8%-10% more correct predictions. Furthermore, the specificity of Pcons is significantly higher than for any individual server. From analyzing different input data to Pcons it can be shown that the improvement is mainly attributable to measurement of the similarity between the different models. Pcons is freely accessible for the academic community through the protein structure-prediction metaserver at http://bioinfo.pl/meta/.     

Can correct protein models be identified?

The ability to separate correct models of protein structures from less correct models is of the greatest importance for protein structure prediction methods. Several studies have examined the ability of different types of energy function to detect the native, or native-like, protein structure from a large set of decoys. In contrast to earlier studies, we examine here the ability to detect models that only show limited structural similarity to the native structure. These correct models are defined by the existence of a fragment that shows significant similarity between this model and the native structure. It has been shown that the existence of such fragments is useful for comparing the performance between different fold recognition methods and that this performance correlates well with performance in fold recognition. We have developed ProQ, a neural-network-based method to predict the quality of a protein model that extracts structural features, such as frequency of atom-atom contacts, and predicts the quality of a model, as measured either by LGscore or MaxSub. We show that ProQ performs at least as well as other measures when identifying the native structure and is better at the detection of correct models. This performance is maintained over several different test sets. ProQ can also be combined with the Pcons fold recognition predictor (Pmodeller) to increase its performance, with the main advantage being the elimination of a few high-scoring incorrect models. Pmodeller was successful in CASP5 and results from the latest LiveBench, LiveBench-6, indicating that Pmodeller has a higher specificity than Pcons alone.     

Identification of correct regions in protein models using structural, alignment, and consensus information

In this study we present two methods to predict the local quality of a protein model: ProQres and ProQprof. ProQres is based on structural features that can be calculated from a model, while ProQprof uses alignment information and can only be used if the model is created from an alignment. In addition, we also propose a simple approach based on local consensus, Pcons-local. We show that all these methods perform better than state-of-the-art methodologies and that, when applicable, the consensus approach is by far the best approach to predict local structure quality. It was also found that ProQprof performed better than other methods for models based on distant relationships, while ProQres performed best for models based on closer relationship, i.e., a model has to be reasonably good to make a structural evaluation useful. Finally, we show that a combination of ProQprof and ProQres (ProQlocal) performed better than any other nonconsensus method for both high- and low-quality models. Additional information and Web servers are available at: http://www.sbc.su.se/~bjorn/ProQ/.     

Ranking predicted protein structures with support vector regression

Protein structure prediction is an important problem of both intellectual and practical interest. Most protein structure prediction approaches generate multiple candidate models first, and then use a scoring function to select the best model among these candidates. In this work, we develop a scoring function using support vector regression (SVR). Both consensus-based features and features from individual structures are extracted from a training data set containing native protein structures and predicted structural models submitted to CASP5 and CASP6. The SVR learns a scoring function that is a linear combination of these features. We test this scoring function on two data sets. First, when used to rank server models submitted to CASP7, the SVR score selects predictions that are comparable to the best performing server in CASP7, Zhang-Server, and significantly better than all the other servers. Even if the SVR score is not allowed to select Zhang-Server models, the SVR score still selects predictions that are significantly better than all the other servers. In addition, the SVR is able to select significantly better models and yield significantly better Pearson correlation coefficients than the two best Quality Assessment groups in CASP7, QA556 (LEE), and QA634 (Pcons). Second, this work aims to improve the ability of the Robetta server to select best models, and hence we evaluate the performance of the SVR score on ranking the Robetta server template-based models for the CASP7 targets. The SVR selects significantly better models than the Robetta K*Sync consensus alignment score.     

基于堆积协变信息与最小自由能预测含伪结的RNA二级结构

RNA伪结预测是RNA研究的一个难点问题。文中提出一种基于堆积协变信息与最小自由能的RNA伪结预测方法。该方法使用已知结构的RNA比对序列(ClustalW比对和结构比对)测试此方法, 侧重考虑相邻碱基对之间相互作用形成的堆积协变信息, 并结合最小自由能方法对碱基配对综合评分, 通过逐步迭代求得含伪结的RNA二级结构。结果表明, 此方法能正确预测伪结, 其平均敏感性和特异性优于参考算法, 并且结构比对的预测性能比ClustalW比对的预测性能更加稳定。文中同时讨论了不同协变信息权重因子对预测性能的影响, 发现权重因子比值在l1: l2=5:1时, 预测性能达到最优。     

Improved predictions by Pcons.net using multiple templates

Multiple templates can often be used to build more accurate homology models than models built from a single template. Here we introduce PconsM, an automated protocol that uses multiple templates to build protein models. PconsM has been among the top-performing methods in the recent CASP experiments and consistently perform better than the single template models used in Pcons.net. In particular for the easier targets with many alternative templates with a high degree of sequence identity, quality is readily improved with a few percentages over the highest ranked model built on a single template. PconsM is available as an additional pipeline within the Pcons.net protein structure prediction server. AVAILABILITY AND IMPLEMENTATION: PconsM is freely available from http://pcons.net/.     

Fold recognition by combining sequence profiles derived from evolution and from depth-dependent structural alignment of fragments

Recognizing structural similarity without significant sequence identity has proved to be a challenging task. Sequence-based and structure-based methods as well as their combinations have been developed. Here, we propose a fold-recognition method that incorporates structural information without the need of sequence-to-structure threading. This is accomplished by generating sequence profiles from protein structural fragments. The structure-derived sequence profiles allow a simple integration with evolution-derived sequence profiles and secondary-structural information for an optimized alignment by efficient dynamic programming. The resulting method (called SP(3)) is found to make a statistically significant improvement in both sensitivity of fold recognition and accuracy of alignment over the method based on evolution-derived sequence profiles alone (SP) and the method based on evolution-derived sequence profile and secondary structure profile (SP(2)). SP(3) was tested in SALIGN benchmark for alignment accuracy and Lindahl, PROSPECTOR 3.0, and LiveBench 8.0 benchmarks for remote-homology detection and model accuracy. SP(3) is found to be the most sensitive and accurate single-method server in all benchmarks tested where other methods are available for comparison (although its results are statistically indistinguishable from the next best in some cases and the comparison is subjected to the limitation of time-dependent sequence and/or structural library used by different methods.). In LiveBench 8.0, its accuracy rivals some of the consensus methods such as ShotGun-INBGU, Pmodeller3, Pcons4, and ROBETTA. SP(3) fold-recognition server is available on http://theory.med.buffalo.edu.     

Microsporidian Encephalitozoon cuniculi, a unicellular eukaryote with an unusual chromosomal dispersion of ribosomal genes and a LSU rRNA reduced to the universal core.

Microsporidia are eukaryotic parasites lacking mitochondria, the ribosomes of which present prokaryote-like features. In order to better understand the structural evolution of rRNA molecules in microsporidia, the 5S and rDNA genes were investigated in Encephalitozoon cuniculi . The genes are not in close proximity. Non-tandemly arranged rDNA units are on every one of the 11 chromosomes. Such a dispersion is also shown in two other Encephalitozoon species. Sequencing of the 5S rRNA coding region reveals a 120 nt long RNA which folds according to the eukaryotic consensus structural shape. In contrast, the LSU rRNA molecule is greatly reduced in length (2487 nt). This dramatic shortening is essentially due to truncation of divergent domains, most of them being removed. Most variable stems of the conserved core are also deleted, reducing the LSU rRNA to only those structural features preserved in all living cells. This suggests that the E.cuniculi LSU rRNA performs only the basic mechanisms of translation. LSU rRNA phylogenetic analysis with the BASEML program favours a relatively recent origin of the fast evolving microsporidian lineage. Therefore, the prokaryote-like ribosomal features, such as the absence of ITS2, may be derived rather than primitive characters.     

MAMMOTH (matching molecular models obtained from theory): an automated method for model comparison

Advances in structural genomics and protein structure prediction require the design of automatic, fast, objective, and well benchmarked methods capable of comparing and assessing the similarity of low-resolution three-dimensional structures, via experimental or theoretical approaches. Here, a new method for sequence-independent structural alignment is presented that allows comparison of an experimental protein structure with an arbitrary low-resolution protein tertiary model. The heuristic algorithm is given and then used to show that it can describe random structural alignments of proteins with different folds with good accuracy by an extreme value distribution. From this observation, a structural similarity score between two proteins or two different conformations of the same protein is derived from the likelihood of obtaining a given structural alignment by chance. The performance of the derived score is then compared with well established, consensus manual-based scores and data sets. We found that the new approach correlates better than other tools with the gold standard provided by a human evaluator. Timings indicate that the algorithm is fast enough for routine use with large databases of protein models. Overall, our results indicate that the new program (MAMMOTH) will be a good tool for protein structure comparisons in structural genomics applications. MAMMOTH is available from our web site at http://physbio.mssm.edu/~ortizg/.     

Structure prediction meta server

SUMMARY: The Structure Prediction Meta Server offers a convenient way for biologists to utilize various high quality structure prediction servers available worldwide. The meta server translates the results obtained from remote services into uniform format, which are consequently used to request a jury prediction from a remote consensus server Pcons. AVAILABILITY: The structure prediction meta server is freely available at http://BioInfo.PL/meta/, some remote servers have however restrictions for non-academic users, which are respected by the meta server. SUPPLEMENTARY INFORMATION: Results of several sessions of the CAFASP and LiveBench programs for assessment of performance of fold-recognition servers carried out via the meta server are available at http://BioInfo.PL/services.html.     

APDB: a web server to evaluate the accuracy of sequence alignments using structural information

The APDB webserver uses structural information to evaluate the alignment of sequences with known structures. It returns a score correlated to the overall alignment accuracy as well as a local evaluation. Any sequence alignment can be analyzed with APDB provided it includes at least two proteins with known structures. Sequences without a known structure are simply ignored and do not contribute to the scoring procedure. AVAILABILITY: APDB is part of the T-Coffee suite of tools for alignment analysis, it is available on www.tcoffee.org. A stand-alone version of the package is also available as a freeware open source from the same address.     

Combining evolutionary and structural information for local protein structure prediction

We study the effects of various factors in representing and combining evolutionary and structural information for local protein structural prediction based on fragment selection. We prepare databases of fragments from a set of non-redundant protein domains. For each fragment, evolutionary information is derived from homologous sequences and represented as estimated effective counts and frequencies of amino acids (evolutionary frequencies) at each position. Position-specific amino acid preferences called structural frequencies are derived from statistical analysis of discrete local structural environments in database structures. Our method for local structure prediction is based on ranking and selecting database fragments that are most similar to a target fragment. Using secondary structure type as a local structural property, we test our method in a number of settings. The major findings are: (1) the COMPASS-type scoring function for fragment similarity comparison gives better prediction accuracy than three other tested scoring functions for profile-profile comparison. We show that the COMPASS-type scoring function can be derived both in the probabilistic framework and in the framework of statistical potentials. (2) Using the evolutionary frequencies of database fragments gives better prediction accuracy than using structural frequencies. (3) Finer definition of local environments, such as including more side-chain solvent accessibility classes and considering the backbone conformations of neighboring residues, gives increasingly better prediction accuracy using structural frequencies. (4) Combining evolutionary and structural frequencies of database fragments, either in a linear fashion or using a pseudocount mixture formula, results in improvement of prediction accuracy. Combination at the log-odds score level is not as effective as combination at the frequency level. This suggests that there might be better ways of combining sequence and structural information than the commonly used linear combination of log-odds scores. Our method of fragment selection and frequency combination gives reasonable results of secondary structure prediction tested on 56 CASP5 targets (average SOV score 0.77), suggesting that it is a valid method for local protein structure prediction. Mixture of predicted structural frequencies and evolutionary frequencies improve the quality of local profile-to-profile alignment by COMPASS.     

Probabilistic consensus scoring improves tandem mass spectrometry peptide identification

Database search is a standard technique for identifying peptides from their tandem mass spectra. To increase the number of correctly identified peptides, we suggest a probabilistic framework that allows the combination of scores from different search engines into a joint consensus score. Central to the approach is a novel method to estimate scores for peptides not found by an individual search engine. This approach allows the estimation of p-values for each candidate peptide and their combination across all search engines. The consensus approach works better than any single search engine across all different instrument types considered in this study. Improvements vary strongly from platform to platform and from search engine to search engine. Compared to the industry standard MASCOT, our approach can identify up to 60% more peptides. The software for consensus predictions is implemented in C++ as part of OpenMS, a software framework for mass spectrometry. The source code is available in the current development version of OpenMS and can easily be used as a command line application or via a graphical pipeline designer TOPPAS.     

No BLUE among phylogenetic estimators

 Multivariate analysis is a branch of statistics that successfully exploits the powerful tools of linear algebra to obtain a fairly comprehensive theory of estimation. The purpose of this paper is to explore to what extent a linear theory of estimation can be developed in the context of coalescent models used in the analysis of DNA polymorphism. We consider a large class of coalescent models, of which the neutral infinite sites model is one example. In the process, we discover several limitations of linear estimators that are quite distinct from those in the classical theory. In particular, we prove that there does not exist a uniformly BLUE (best linear unbiased estimator) for the scaled mutation parameter, under the assumptions of the neutral model of evolution. In fact, we show that no linear estimator performs uniformly better than the Watterson (1975) method based on the total number of segregating sites. For certain coalescent models, the segregating-sites estimator is actually optimal. The general conclusion is the following. If genealogical information is useful for estimating the rate of evolution, then there is no optimal linear method. If there is an optimal linear method, then no information other than the total number of segregating sites is needed. Received: 29 July 1998 / Revised version: 9 October 1998     

Identification of related proteins on family, superfamily and fold level

Proteins might have considerable structural similarities even when no evolutionary relationship of their sequences can be detected. This property is often referred to as the proteins sharing only a "fold". Of course, there are also sequences of common origin in each fold, called a "superfamily", and in them groups of sequences with clear similarities, designated "family". Developing algorithms to reliably identify proteins related at any level is one of the most important challenges in the fast growing field of bioinformatics today. However, it is not at all certain that a method proficient at finding sequence similarities performs well at the other levels, or vice versa.Here, we have compared the performance of various search methods on these different levels of similarity. As expected, we show that it becomes much harder to detect proteins as their sequences diverge. For family related sequences the best method gets 75% of the top hits correct. When the sequences differ but the proteins belong to the same superfamily this drops to 29%, and in the case of proteins with only fold similarity it is as low as 15%. We have made a more complete analysis of the performance of different algorithms than earlier studies, also including threading methods in the comparison. Using this method a more detailed picture emerges, showing multiple sequence information to improve detection on the two closer levels of relationship. We have also compared the different methods of including this information in prediction algorithms.For lower specificities, the best scheme to use is a linking method connecting proteins through an intermediate hit. For higher specificities, better performance is obtained by PSI-BLAST and some procedures using hidden Markov models. We also show that a threading method, THREADER, performs significantly better than any other method at fold recognition.     

Secondary structure prediction with support vector machines

MOTIVATION: A new method that uses support vector machines (SVMs) to predict protein secondary structure is described and evaluated. The study is designed to develop a reliable prediction method using an alternative technique and to investigate the applicability of SVMs to this type of bioinformatics problem. METHODS: Binary SVMs are trained to discriminate between two structural classes. The binary classifiers are combined in several ways to predict multi-class secondary structure. RESULTS: The average three-state prediction accuracy per protein (Q(3)) is estimated by cross-validation to be 77.07 +/- 0.26% with a segment overlap (Sov) score of 73.32 +/- 0.39%. The SVM performs similarly to the 'state-of-the-art' PSIPRED prediction method on a non-homologous test set of 121 proteins despite being trained on substantially fewer examples. A simple consensus of the SVM, PSIPRED and PROFsec achieves significantly higher prediction accuracy than the individual methods.     

Combination of text-mining algorithms increases the performance

MOTIVATION: Recently, several information extraction systems have been developed to retrieve relevant information out of biomedical text. However, these methods represent individual efforts. In this paper, we show that by combining different algorithms and their outcome, the results improve significantly. For this reason, CONAN has been created, a system which combines different programs and their outcome. Its methods include tagging of gene/protein names, finding interaction and mutation data, tagging of biological concepts and linking to MeSH and Gene Ontology terms. RESULTS: In this paper, we will present data that show that combining different text-mining algorithms significantly improves the results. Not only is CONAN a full-scale approach that will ultimately cover all of PubMed/MEDLINE, we also show that this universality has no effect on quality: our system performs as well as or better than existing systems. AVAILABILITY: The LDD corpus presented is available by request to the author. The system will be available shortly. For information and updates on CONAN please visit http://www.cs.uu.nl/people/rainer/conan.html.     

QMEANclust: estimation of protein model quality by combining a composite scoring function with structural density information

Background  

The selection of the most accurate protein model from a set of alternatives is a crucial step in protein structure prediction both in template-based and ab initio approaches. Scoring functions have been developed which can either return a quality estimate for a single model or derive a score from the information contained in the ensemble of models for a given sequence. Local structural features occurring more frequently in the ensemble have a greater probability of being correct. Within the context of the CASP experiment, these so called consensus methods have been shown to perform considerably better in selecting good candidate models, but tend to fail if the best models are far from the dominant structural cluster. In this paper we show that model selection can be improved if both approaches are combined by pre-filtering the models used during the calculation of the structural consensus.     

Linkage analysis using sex-specific recombination fractions with GENEHUNTER-MODSCORE

MOTIVATION: Sex-specific marker maps have become increasingly available. We have implemented the usage of sex-specific recombination frequencies in the GENEHUNTER-MODSCORE program that performs multipoint linkage analysis. Furthermore, we have devised a consistent method to choose the combinations of male and female genetic positions at which linkage scores should be calculated. Marker coordinates can be read automatically from publicly available genetic maps. RESULTS: In a MOD-score analysis of the COGA dataset provided for Genetic Analysis Workshop 14, the highest linkage peak on chromosome 1 further increases when using sex-specific maps, while some smaller peaks are decreased. Simulations confirm that the MOD score can be biased when a sex-averaged instead of the correct sex-specific map is employed. This shows that an adequate modeling of the female:male ratio of genetic distances is important, especially for complex traits. AVAILABILITY: The new version of GENEHUNTER-MODSCORE can be downloaded from the following website: http://www.staff.uni-marburg.de/~strauchk/software.html     

DDBASE2.0: updated domain database with improved identification of structural domains

MOTIVATION: Although many methods are available for the identification of structural domains from protein three-dimensional structures, accurate definition of protein domains and the curation of such data for a large number of proteins are often possible only after manual intervention. The availability of domain definitions for protein structural entries is useful for the sequence analysis of aligned domains, structure comparison, fold recognition procedures and understanding protein folding, domain stability and flexibility. RESULTS: We have improved our method of domain identification starting from the concept of clustering secondary structural elements, but with an intention of reducing the number of discontinuous segments in identified domains. The results of our modified and automatic approach have been compared with the domain definitions from other databases. On a test data set of 55 proteins, this method acquires high agreement (88%) in the number of domains with the crystallographers' definition and resources such as SCOP, CATH, DALI, 3Dee and PDP databases. This method also obtains 98% overlap score with the other resources in the definition of domain boundaries of the 55 proteins. We have examined the domain arrangements of 4592 non-redundant protein chains using the improved method to include 5409 domains leading to an update of the structural domain database. AVAILABILITY: The latest version of the domain database and online domain identification methods are available from http://www.ncbs.res.in/~faculty/mini/ddbase/ddbase.html Supplementary information: http://www.ncbs.res.in/~faculty/mini/ddbase/supplementary/supplementary.html