An iterative refinement algorithm for consistency based multiple structural alignment methods

MOTIVATION: Multiple STructural Alignment (MSTA) provides valuable information for solving problems such as fold recognition. The consistency-based approach tries to find conflict-free subsets of alignments from a pre-computed all-to-all Pairwise Alignment Library (PAL). If large proportions of conflicts exist in the library, consistency can be hard to get. On the other hand, multiple structural superposition has been used in many MSTA methods to refine alignments. However, multiple structural superposition is dependent on alignments, and a superposition generated based on erroneous alignments is not guaranteed to be the optimal superposition. Correcting errors after making errors is not as good as avoiding errors from the beginning. Hence it is important to refine the pairwise library to reduce the number of conflicts before any consistency-based assembly. RESULTS: We present an algorithm, Iterative Refinement of Induced Structural alignment (IRIS), to refine the PAL. A new measurement for the consistency of a library is also proposed. Experiments show that our algorithm can greatly improve T-COFFEE performance for less consistent pairwise alignment libraries. The final multiple alignment outperforms most state-of-the-art MSTA algorithms at assembling 15 transglycosidases. Results on three other benchmarks showed that the algorithm consistently improves multiple alignment performance. AVAILABILITY: The C++ code of the algorithm is available upon request.     

Progressive structure-based alignment of homologous proteins: Adopting sequence comparison strategies

Comparison of multiple protein structures has a broad range of applications in the analysis of protein structure, function and evolution. Multiple structure alignment tools (MSTAs) are necessary to obtain a simultaneous comparison of a family of related folds. In this study, we have developed a method for multiple structure comparison largely based on sequence alignment techniques. A widely used Structural Alphabet named Protein Blocks (PBs) was used to transform the information on 3D protein backbone conformation as a 1D sequence string. A progressive alignment strategy similar to CLUSTALW was adopted for multiple PB sequence alignment (mulPBA). Highly similar stretches identified by the pairwise alignments are given higher weights during the alignment. The residue equivalences from PB based alignments are used to obtain a three dimensional fit of the structures followed by an iterative refinement of the structural superposition. Systematic comparisons using benchmark datasets of MSTAs underlines that the alignment quality is better than MULTIPROT, MUSTANG and the alignments in HOMSTRAD, in more than 85% of the cases. Comparison with other rigid-body and flexible MSTAs also indicate that mulPBA alignments are superior to most of the rigid-body MSTAs and highly comparable to the flexible alignment methods.     

Development and validation of a consistency based multiple structure alignment algorithm

SUMMARY: We introduce an algorithm that uses the information gained from simultaneous consideration of an entire group of related proteins to create multiple structure alignments (MSTAs). Consistency-based alignment (CBA) first harnesses the information contained within regions that are consistently aligned among a set of pairwise superpositions in order to realign pairs of proteins through both global and local refinement methods. It then constructs a multiple alignment that is maximally consistent with the improved pairwise alignments. We validate CBA's alignments by assessing their accuracy in regions where at least two of the aligned structures contain the same conserved sequence motif. RESULTS: CBA correctly aligns well over 90% of motif residues in superpositions of proteins belonging to the same family or superfamily, and it outperforms a number of previously reported MSTA algorithms.     

Combining many multiple alignments in one improved alignment

MOTIVATION: The fact that the multiple sequence alignment problem is of high complexity has led to many different heuristic algorithms attempting to find a solution in what would be considered a reasonable amount of computation time and space. Very few of these heuristics produce results that are guaranteed always to lie within a certain distance of an optimal solution (given a measure of quality, e.g. parsimony). Most practical heuristics cannot guarantee this, but nevertheless perform well for certain cases. An alignment, obtained with one of these heuristics and with a bad overall score, is not unusable though, it might contain important information on how substrings should be aligned. This paper presents a method that extracts qualitatively good sub-alignments from a set of multiple alignments and combines these into a new, often improved alignment. The algorithm is implemented as a variant of the traditional dynamic programming technique. RESULTS: An implementation of ComAlign (the algorithm that combines multiple alignments) has been run on several sets of artificially generated sequences and a set of 5S RNA sequences. To assess the quality of the alignments obtained, the results have been compared with the output of MSA 2.1 (Gupta et al., Proceedings of the Sixth Annual Symposium on Combinatorial Pattern Matching, 1995; Kececioglu et al., http://www.techfak.uni-bielefeld. de/bcd/Lectures/kececioglu.html, 1995). In all cases, ComAlign was able to produce a solution with a score comparable to the solution obtained by MSA. The results also show that ComAlign actually does combine parts from different alignments and not just select the best of them. AVAILABILITY: The C source code (a Smalltalk version is being worked on) of ComAlign and the other programs that have been implemented in this context are free and available on WWW (http://www.daimi.au.dk/ ?caprani). CONTACT: klaus@bucka-lassen.dk; jotun@pop.bio.au.dk;ocaprani@daimi.au.dk     

Mocca: semi-automatic method for domain hunting

MOTIVATION: Multiple OCCurrences Analysis (Mocca) is a new method for repeat extraction. It is based on the T-Coffee package (Notredame et al., JMB, 302, 205-217, 2000). Given a sequence or a set of sequences, and a library of local alignments, Mocca extracts every segment of sequence homologous to a pre-specified master. The implementation is meant for domain hunting and makes it fast and easy to test for new boundaries or extend known repeats in an interactive manner. Mocca is designed to deal with highly divergent protein repeats (less than 30% amino acid identity) of more than 30 amino acids.     

Using the T-Coffee package to build multiple sequence alignments of protein, RNA, DNA sequences and 3D structures

T-Coffee (Tree-based consistency objective function for alignment evaluation) is a versatile multiple sequence alignment (MSA) method suitable for aligning most types of biological sequences. The main strength of T-Coffee is its ability to combine third party aligners and to integrate structural (or homology) information when building MSAs. The series of protocols presented here show how the package can be used to multiply align proteins, RNA and DNA sequences. The protein section shows how users can select the most suitable T-Coffee mode for their data set. Detailed protocols include T-Coffee, the default mode, M-Coffee, a meta version able to combine several third party aligners into one, PSI (position-specific iterated)-Coffee, the homology extended mode suitable for remote homologs and Expresso, the structure-based multiple aligner. We then also show how the T-RMSD (tree based on root mean square deviation) option can be used to produce a functionally informative structure-based clustering. RNA alignment procedures are described for using R-Coffee, a mode able to use predicted RNA secondary structures when aligning RNA sequences. DNA alignments are illustrated with Pro-Coffee, a multiple aligner specific of promoter regions. We also present some of the many reformatting utilities bundled with T-Coffee. The package is an open-source freeware available from http://www.tcoffee.org/.     

Improving the quality of twilight-zone alignments

Several recent publications illustrated advantages of using sequence profiles in recognizing distant homologies between proteins. At the same time, the practical usefulness of distant homology recognition depends not only on the sensitivity of the algorithm, but also on the quality of the alignment between a prediction target and the template from the database of known proteins. Here, we study this question for several supersensitive protein algorithms that were previously compared in their recognition sensitivity (Rychlewski et al., 2000). A database of protein pairs with similar structures, but low sequence similarity is used to rate the alignments obtained with several different methods, which included sequence-sequence, sequence-profile, and profile-profile alignment methods. We show that incorporation of evolutionary information encoded in sequence profiles into alignment calculation methods significantly increases the alignment accuracy, bringing them closer to the alignments obtained from structure comparison. In general, alignment quality is correlated with recognition and alignment score significance. For every alignment method, alignments with statistically significant scores correlate with both correct structural templates and good quality alignments. At the same time, average alignment lengths differ in various methods, making the comparison between them difficult. For instance, the alignments obtained by FFAS, the profile-profile alignment algorithm developed in our group are always longer that the alignments obtained with the PSI-BLAST algorithms. To address this problem, we develop methods to truncate or extend alignments to cover a specified percentage of protein lengths. In most cases, the elongation of the alignment by profile-profile methods is reasonable, adding fragments of similar structure. The examples of erroneous alignment are examined and it is shown that they can be identified based on the model quality.     

HOMSTRAD: adding sequence information to structure-based alignments of homologous protein families

summary: We describe an extension to the Homologous Structure Alignment Database (HOMSTRAD; Mizuguchi et al., Protein Sci., 7, 2469-2471, 1998a) to include homologous sequences derived from the protein families database Pfam (Bateman et al., Nucleic Acids Res., 28, 263-266, 2000). HOMSTRAD is integrated with the server FUGUE (Shi et al., submitted, 2001) for recognition and alignment of homologues, benefitting from the combination of abundant sequence information and accurate structure-based alignments. AVAILABILITY The HOMSTRAD database is available at: http://www-cryst.bioc.cam.ac.uk/homstrad/. Query sequences can be submitted to the homology recognition/alignment server FUGUE at: http://www-cryst.bioc.cam.ac.uk/fugue/.     

DBAli: a database of protein structure alignments

SUMMARY: The DBAli database includes approximately 35000 alignments of pairs of protein structures from SCOP (Lo Conte et al., Nucleic Acids Res., 28, 257-259, 2000) and CE (Shindyalov and Bourne, Protein Eng., 11, 739-747, 1998). DBAli is linked to several resources, including Compare3D (Shindyalov and Bourne, http://www.sdsc.edu/pb/software.htm, 1999) and ModView (Ilyin and Sali, http://guitar.rockefeller.edu/ModView/, 2001) for visualizing sequence alignments and structure superpositions. A flexible search of DBAli by protein sequence and structure properties allows construction of subsets of alignments suitable for a number of applications, such as benchmarking of sequence-sequence and sequence-structure alignment methods under a variety of conditions. AVAILABILITY: http://guitar.rockefeller.edu/DBAli/     

Quality assessment of multiple alignment programs

A renewed interest in the multiple sequence alignment problem has given rise to several new algorithms. In contrast to traditional progressive methods, computationally expensive score optimization strategies are now predominantly employed. We systematically tested four methods (Poa, Dialign, T-Coffee and ClustalW) for the speed and quality of their alignments. As test sequences we used structurally derived alignments from BAliBASE and synthetic alignments generated by Rose. The tests included alignments of variable numbers of domains embedded in random spacer sequences. Overall, Dialign was the most accurate in cases with low sequence identity, while T-Coffee won in cases with high sequence identity. The fast Poa algorithm was almost as accurate, while ClustalW could compete only in strictly global cases with high sequence similarity.     

Multiple sequence alignments

Multiple sequence alignments are very widely used in all areas of DNA and protein sequence analysis. The main methods that are still in use are based on 'progressive alignment' and date from the mid to late 1980s. Recently, some dramatic improvements have been made to the methodology with respect either to speed and capacity to deal with large numbers of sequences or to accuracy. There have also been some practical advances concerning how to combine three-dimensional structural information with primary sequences to give more accurate alignments, when structures are available.     

Predicting reliable regions in protein sequence alignments

MOTIVATION: Protein sequence alignments have a myriad of applications in bioinformatics, including secondary and tertiary structure prediction, homology modeling, and phylogeny. Unfortunately, all alignment methods make mistakes, and mistakes in alignments often yield mistakes in their application. Thus, a method to identify and remove suspect alignment positions could benefit many areas in protein sequence analysis. RESULTS: We tested four predictors of alignment position reliability, including near-optimal alignment information, column score, and secondary structural information. We validated each predictor against a large library of alignments, removing positions predicted as unreliable. Near-optimal alignment information was the best predictor, removing 70% of the substantially-misaligned positions and 58% of the over-aligned positions, while retaining 86% of those aligned accurately.     

FAST: Fourier transform based algorithms for significance testing of ungapped multiple alignments

SUMMARY: As was shown in Nagarajan et al. (2005), commonly used approximations for assessing the significance of multiple alignments can be be very inaccurate. To address this, we present here the FAST package, an open-source collection of programs and libraries for efficiently and reliably computing the significance of ungapped local alignments. We also describe other potential applications in Bioinformatics where these programs can be adapted for significance testing. AVAILABILITY: The FAST package includes C++ implementations of various algorithms that can be used as stand-alone programs or as a library of subroutines. The package and a web-server for some of the programs are available at www.cs.cornell.edu/~keich/FAST.     

On the functional significance of electron density protein structure alignments

Electron density protein alignments are analyzed in terms of their underlying similarity measure, the density overlap. These alignments are conceptually unrelated to biochemical structural elements and, therefore, are appropriate in structure-only similarity studies. The analysis is focused on the low sequence similarity subset of protein domains. A remarkable association is found between simple, density overlap measures and the expert designed Structural Classification of Proteins (SCOP) for which functional and evolutive analogies prevail. The association found validates the functional significance of electron density alignments.     

HMM-Kalign: a tool for generating sub-optimal HMM alignments

Recent development of strategies using multiple sequence alignments (MSA) or profiles to detect remote homologies between proteins has led to a significant increase in the number of proteins whose structures can be generated by comparative modeling methods. However, prediction of the optimal alignment between these highly divergent homologous proteins remains a difficult issue. We present a tool based on a generalized Viterbi algorithm that generates optimal and sub-optimal alignments between a sequence and a Hidden Markov Model. The tool is implemented as a new function within the HMMER package called hmmkalign.     

Multiple sequence alignment with arbitrary gap costs: computing an optimal solution using polyhedral combinatorics

Multiple sequence alignment is one of the dominant problems in computational molecular biology. Numerous scoring functions and methods have been proposed, most of which result in NP-hard problems. In this paper we propose for the first time a general formulation for multiple alignment with arbitrary gap-costs based on an integer linear program (ILP). In addition we describe a branch-and-cut algorithm to effectively solve the ILP to optimality. We evaluate the performances of our approach in terms of running time and quality of the alignments using the BAliBase database of reference alignments. The results show that our implementation ranks amongst the best programs developed so far.     

3DCoffee: combining protein sequences and structures within multiple sequence alignments

Most bioinformatics analyses require the assembly of a multiple sequence alignment. It has long been suspected that structural information can help to improve the quality of these alignments, yet the effect of combining sequences and structures has not been evaluated systematically. We developed 3DCoffee, a novel method for combining protein sequences and structures in order to generate high-quality multiple sequence alignments. 3DCoffee is based on TCoffee version 2.00, and uses a mixture of pairwise sequence alignments and pairwise structure comparison methods to generate multiple sequence alignments. We benchmarked 3DCoffee using a subset of HOMSTRAD, the collection of reference structural alignments. We found that combining TCoffee with the threading program Fugue makes it possible to improve the accuracy of our HOMSTRAD dataset by four percentage points when using one structure only per dataset. Using two structures yields an improvement of ten percentage points. The measures carried out on HOM39, a HOMSTRAD subset composed of distantly related sequences, show a linear correlation between multiple sequence alignment accuracy and the ratio of number of provided structure to total number of sequences. Our results suggest that in the case of distantly related sequences, a single structure may not be enough for computing an accurate multiple sequence alignment.     

Tcoffee@igs: A web server for computing,evaluating and combining multiple sequence alignments

This paper presents Tcoffee@igs, a new server provided to the community by Hewlet Packard computers and the Centre National de la Recherche Scientifique. This server is a web-based tool dedicated to the computation, the evaluation and the combination of multiple sequence alignments. It uses the latest version of the T-Coffee package. Given a set of unaligned sequences, the server returns an evaluated multiple sequence alignment and the associated phylogenetic tree. This server also makes it possible to evaluate the local reliability of an existing alignment and to combine several alternative multiple alignments into a single new one. Tcoffee@igs can be used for aligning protein, RNA or DNA sequences. Datasets of up to 100 sequences (2000 residues long) can be processed. The server and its documentation are available from: http://igs-server.cnrs-mrs.fr/Tcoffee/.     

MUSCLE: multiple sequence alignment with high accuracy and high throughput

We describe MUSCLE, a new computer program for creating multiple alignments of protein sequences. Elements of the algorithm include fast distance estimation using kmer counting, progressive alignment using a new profile function we call the log-expectation score, and refinement using tree-dependent restricted partitioning. The speed and accuracy of MUSCLE are compared with T-Coffee, MAFFT and CLUSTALW on four test sets of reference alignments: BAliBASE, SABmark, SMART and a new benchmark, PREFAB. MUSCLE achieves the highest, or joint highest, rank in accuracy on each of these sets. Without refinement, MUSCLE achieves average accuracy statistically indistinguishable from T-Coffee and MAFFT, and is the fastest of the tested methods for large numbers of sequences, aligning 5000 sequences of average length 350 in 7 min on a current desktop computer. The MUSCLE program, source code and PREFAB test data are freely available at http://www.drive5. com/muscle.     

R-Coffee: a method for multiple alignment of non-coding RNA

R-Coffee is a multiple RNA alignment package, derived from T-Coffee, designed to align RNA sequences while exploiting secondary structure information. R-Coffee uses an alignment-scoring scheme that incorporates secondary structure information within the alignment. It works particularly well as an alignment improver and can be combined with any existing sequence alignment method. In this work, we used R-Coffee to compute multiple sequence alignments combining the pairwise output of sequence aligners and structural aligners. We show that R-Coffee can improve the accuracy of all the sequence aligners. We also show that the consistency-based component of T-Coffee can improve the accuracy of several structural aligners. R-Coffee was tested on 388 BRAliBase reference datasets and on 11 longer Cmfinder datasets. Altogether our results suggest that the best protocol for aligning short sequences (less than 200 nt) is the combination of R-Coffee with the RNA pairwise structural aligner Consan. We also show that the simultaneous combination of the four best sequence alignment programs with R-Coffee produces alignments almost as accurate as those obtained with R-Coffee/Consan. Finally, we show that R-Coffee can also be used to align longer datasets beyond the usual scope of structural aligners. R-Coffee is freely available for download, along with documentation, from the T-Coffee web site (www.tcoffee.org).