MOTIVATION: Biologically important proteins are often large, multidomain proteins, which are difficult to characterize by high-throughput experimental methods. Efficient domain/boundary predictions are thus increasingly required in diverse area of proteomics research for computationally dissecting proteins into readily analyzable domains. RESULTS: We constructed a support vector machine (SVM)-based domain linker predictor, DROP (Domain linker pRediction using OPtimal features), which was trained with 25 optimal features. The optimal combination of features was identified from a set of 3000 features using a random forest algorithm complemented with a stepwise feature selection. DROP demonstrated a prediction sensitivity and precision of 41.3 and 49.4%, respectively. These values were over 19.9% higher than those of control SVM predictors trained with non-optimized features, strongly suggesting the efficiency of our feature selection method. In addition, the mean NDO-Score of DROP for predicting novel domains in seven CASP8 FM multidomain proteins was 0.760, which was higher than any of the 12 published CASP8 DP servers. Overall, these results indicate that the SVM prediction of domain linkers can be improved by identifying optimal features that best distinguish linker from non-linker regions.

SUMMARY: Multiple sequence alignment (MSA) is an important step in comparative sequence analyses. Parallelization is a key technique for reducing the time required for large-scale sequence analyses. The three calculation stages, all-to-all comparison, progressive alignment and iterative refinement, of the MAFFT MSA program were parallelized using the POSIX Threads library. Two natural parallelization strategies (best-first and simple hill-climbing) were implemented for the iterative refinement stage. Based on comparisons of the objective scores and benchmark scores between the two approaches, we selected a simple hill-climbing approach as the default. AVAILABILITY: The parallelized version of MAFFT is available at http://mafft.cbrc.jp/alignment/software/. This version currently supports the Linux operating system only.

MOTIVATION: Functional similarity between proteins is evident at both the sequence and structure levels. SeSAW is a web-based program for identifying functionally or evolutionarily conserved motifs in protein structures by locating sequence and structural similarities, and quantifying these at the level of individual residues. Results can be visualized in 2D, as annotated alignments, or in 3D, as structural superpositions. An example is given for both an experimentally determined query structure and a homology model. AVAILABILITY AND IMPLEMENTATION: The web server is located at http://www.pdbj.org/SeSAW/.

BACKGROUND: Conservation and variation scores are used when evaluating sites in a multiple sequence alignment, in order to identify residues critical for structure or function. A variety of scores are available today but it is not clear how different scores relate to each other. RESULTS: We applied 25 conservation and variation scores to alignments from the Catalytic Site Atlas (CSA). We calculated distances among scores based on correlation coefficients, and constructed a dendrogram of the scores by average linking cluster analysis. The cluster analysis showed that most scores fall into one of two groups--substitution matrix based group and frequency based group respectively. We also evaluated the scores' performance in predicting catalytic sites and found that frequency based scores generally perform best. CONCLUSIONS: Conservation and variation scores can be classified into mainly two large groups. When using a score to predict catalytic sites, frequency based scores that also consider a background distribution are most successful.

The accuracy and scalability of multiple sequence alignment (MSA) of DNAs and proteins have long been and are still important issues in bioinformatics. To rapidly construct a reasonable MSA, we developed the initial version of the MAFFT program in 2002. MSA software is now facing greater challenges in both scalability and accuracy than those of 5 years ago. As increasing amounts of sequence data are being generated by large-scale sequencing projects, scalability is now critical in many situations. The requirement of accuracy has also entered a new stage since the discovery of functional noncoding RNAs (ncRNAs); the secondary structure should be considered for constructing a high-quality alignment of distantly related ncRNAs. To deal with these problems, in 2007, we updated MAFFT to Version 6 with two new techniques: the PartTree algorithm and the Four-way consistency objective function. The former improved the scalability of progressive alignment and the latter improved the accuracy of ncRNA alignment. We review these and other techniques that MAFFT uses and suggest possible future directions of MSA software as a basis of comparative analyses. MAFFT is available at http://align.bmr.kyushu-u.ac.jp/mafft/software/.

BACKGROUND: Structural alignment of RNAs is becoming important, since the discovery of functional non-coding RNAs (ncRNAs). Recent studies, mainly based on various approximations of the Sankoff algorithm, have resulted in considerable improvement in the accuracy of pairwise structural alignment. In contrast, for the cases with more than two sequences, the practical merit of structural alignment remains unclear as compared to traditional sequence-based methods, although the importance of multiple structural alignment is widely recognized. RESULTS: We took a different approach from a straightforward extension of the Sankoff algorithm to the multiple alignments from the viewpoints of accuracy and time complexity. As a new option of the MAFFT alignment program, we developed a multiple RNA alignment framework, X-INS-i, which builds a multiple alignment with an iterative method incorporating structural information through two components: (1) pairwise structural alignments by an external pairwise alignment method such as SCARNA or LaRA and (2) a new objective function, Four-way Consistency, derived from the base-pairing probability of every sub-aligned group at every multiple alignment stage. CONCLUSION: The BRAliBASE benchmark showed that X-INS-i outperforms other methods currently available in the sum-of-pairs score (SPS) criterion. As a basis for predicting common secondary structure, the accuracy of the present method is comparable to or rather higher than those of the current leading methods such as RNA Sampler. The X-INS-i framework can be used for building a multiple RNA alignment from any combination of algorithms for pairwise RNA alignment and base-pairing probability. The source code is available at the webpage found in the Availability and requirements section.

MOTIVATION: To construct a multiple sequence alignment (MSA) of a large number (> approximately 10,000) of sequences, the calculation of a guide tree with a complexity of O(N2) to O(N3), where N is the number of sequences, is the most time-consuming process. RESULTS: To overcome this limitation, we have developed an approximate algorithm, PartTree, to construct a guide tree with an average time complexity of O(N log N). The new MSA method with the PartTree algorithm can align approximately 60,000 sequences in several minutes on a standard desktop computer. The loss of accuracy in MSA caused by this approximation was estimated to be several percent in benchmark tests using Pfam. AVAILABILITY: The present algorithm has been implemented in the MAFFT sequence alignment package (http://align.bmr.kyushu-u.ac.jp/mafft/software/). SUPPLEMENTARY INFORMATION: Supplementary information is available at Bioinformatics online.

BACKGROUND: Structure alignment methods offer the possibility of measuring distant evolutionary relationships between proteins that are not visible by sequence-based analysis. However, the question of how structural differences and similarities ought to be quantified in this regard remains open. In this study we construct a training set of sequence-unique CATH and SCOP domains, from which we develop a scoring function that can reliably identify domains with the same CATH topology and SCOP fold classification. The score is implemented in the ASH structure alignment package, for which the source code and a web service are freely available from the PDBj website http://www.pdbj.org/ASH/. RESULTS: The new ASH score shows increased selectivity and sensitivity compared with values reported for several popular programs using the same test set of 4,298,905 structure pairs, yielding an area of .96 under the receiver operating characteristic (ROC) curve. In addition, weak sequence homologies between similar domains are revealed that could not be detected by BLAST sequence alignment. Also, a subset of domain pairs is identified that exhibit high similarity, even though their CATH and SCOP classification differs. Finally, we show that the ranking of alignment programs based solely on geometric measures depends on the choice of the quality measure. CONCLUSION: ASH shows high selectivity and sensitivity with regard to domain classification, an important step in defining distantly related protein sequence families. Moreover, the CPU cost per alignment is competitive with the fastest programs, making ASH a practical option for large-scale structure classification studies.

MOTIVATION: The computational prediction of protein-protein interactions is currently a major issue in bioinformatics. Recently, a variety of co-evolution-based methods have been investigated toward this goal. In this study, we introduced a partial correlation coefficient as a new measure for the degree of co-evolution between proteins, and proposed its use to predict protein-protein interactions. RESULTS: The accuracy of the prediction by the proposed method was compared with those of the original mirror tree method and the projection method previously developed by our group. We found that the partial correlation coefficient effectively reduces the number of false positives, as compared with other methods, although the number of false negatives increased in the prediction by the partial correlation coefficient. AVAILABILITY: The R script for the prediction of protein-protein interactions reported in this manuscript is available at http://timpani.genome.ad.jp/~parco/

MOTIVATION: The prediction of protein-protein interactions is currently an important issue in bioinformatics. The mirror tree method uses evolutionary information to predict protein-protein interactions. However, it has been recognized that predictions by the mirror tree method lead to many false positives. The incentive of our study was to solve this problem by improving the method of extracting the co-evolutionary information regarding the protein pairs. RESULTS: We developed a novel method to predict protein-protein interactions from co-evolutionary information in the framework of the mirror tree method. The originality is the use of the projection operator to exclude the information about the phylogenetic relationships among the source organisms from the distance matrix. Each distance matrix was transformed into a vector for the operation. The vector is referred to as a 'phylogenetic vector'. We have proposed three ways to extract the phylogenetic information: (1) using the 16S rRNA from the same source organisms as the proteins under consideration, (2) averaging the phylogenetic vectors and (3) analyzing the principal components of the phylogenetic vectors. We examined the performance of the proposed methods to predict interacting protein pairs from Escherichia coli, using experimentally verified data. Our method was successful, and it drastically reduced the number of false positives in the prediction. AVAILABILITY: The R script for the prediction of protein-protein interactions reported in this manuscript is available at http://timpani.genome.ad.jp/~proj/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: The information is also available at the same site as the R script.

BACKGROUND: We introduce GASH, a new, publicly accessible program for structural alignment and superposition. Alignments are scored by the Number of Equivalent Residues (NER), a quantitative measure of structural similarity that can be applied to any structural alignment method. Multiple alignments are optimized by conjugate gradient maximization of the NER score within the genetic algorithm framework. Initial alignments are generated by the program Local ASH, and can be supplemented by alignments from any other program. RESULTS: We compare GASH to DaliLite, CE, and to our earlier program Global ASH on a difficult test set consisting of 3,102 structure pairs, as well as a smaller set derived from the Fischer-Eisenberg set. The extent of alignment crossover, as well as the completeness of the initial set of alignments are examined. The quality of the superpositions is evaluated both by NER and by the number of aligned residues under three different RMSD cutoffs (2,4, and 6A). In addition to the numerical assessment, the alignments for several biologically related structural pairs are discussed in detail. CONCLUSION: Regardless of which criteria is used to judge the superposition accuracy, GASH achieves the best overall performance, followed by DaliLite, Global ASH, and CE. In terms of CPU usage, DaliLite CE and GASH perform similarly for query proteins under 500 residues, but for larger proteins DaliLite is faster than GASH or CE. Both an http interface and a simple object application protocol (SOAP) interface to the GASH program are available at http://www.pdbj.org/GASH/.

UNLABELLED: We constructed a website for inferring a network by applying the graphical Gaussian model, from a large amount of data, including redundant information. The available tools on the website are based on a system, named ASIAN (Automatic System for Inferring A Network), in combination with the two methods in our previous papers, which were designed to analyze gene expression profiles on a genomic scale. One of the remarkable features of the website is its ability to infer a network, concomitant with hierarchical clustering and the following estimation of cluster boundaries. AVAILABILITY: http://eureka.ims.u-tokyo.ac.jp/asian

MOTIVATION: Recent advances in DNA microarray technologies have made it possible to measure the expression levels of thousands of genes simultaneously under different conditions. The data obtained by microarray analyses are called expression profile data. One type of important information underlying the expression profile data is the 'genetic network,' that is, the regulatory network among genes. Graphical Gaussian Modeling (GGM) is a widely utilized method to infer or test relationships among a plural of variables. RESULTS: In this study, we developed a method combining the cluster analysis with GGM for the inference of the genetic network from the expression profile data. The expression profile data of 2467 Saccharomyces cerevisiae genes measured under 79 different conditions (Eisen et al., PROC: Natl Acad. Sci. USA, 95, 14683-14868, 1998) were used for this study. At first, the 2467 genes were classified into 34 clusters by a cluster analysis, as a preprocessing for GGM. Then, the expression levels of the genes in each cluster were averaged for each condition. The averaged expression profile data of 34 clusters were subjected to GGM, and a partial correlation coefficient matrix was obtained as a model of the genetic network of S. cerevisiae. The accuracy of the inferred network was examined by the agreement of our results with the cumulative results of experimental studies.

MOTIVATION: Gene expression profile data are rapidly accumulating due to advances in microarray techniques. The abundant data are analyzed by clustering procedures to extract the useful information about the genes inherent in the data. In the clustering analyses, the systematic determination of the boundaries of gene clusters, instead of by visual inspection and biological knowledge, still remains challenging. RESULTS: We propose a statistical procedure to estimate the number of clusters in the hierarchical clustering of the expression profiles. Following the hierarchical clustering, the statistical property of the profiles at the node in the dendrogram is evaluated by a statistics-based value: the variance inflation factor in the multiple regression analysis. The evaluation leads to an automatic determination of the cluster boundaries without any additional analyses and any biological knowledge of the measured genes. The performance of the present procedure is demonstrated on the profiles of 2467 yeast genes, with very promising results. AVAILABILITY: A set of programs will be electronically sent upon request. CONTACT: [email protected]; [email protected]

MOTIVATION: Data on the entire structures of organelle and bacterial genomes, most of which are known to be circular, have accumulated at a rapid pace. This information enables us to utilize the locations of homologous gene pairs for measuring the dissimilarity between complete genomic structures. RESULTS: A macroscopic distance is presented for comparing circular genomes from their overall structures, on the basis of the locations of two pairs of homologous genes on the compared genomes. The novel aspect of our method is that the comparison between the genomes automatically reveals a relationship based on the information on all gene locations, by incorporating the mobility of each gene, which includes not only the gene order, but also the relative location between gene pairs. The plausibility of the newly defined distances is evaluated by means of 44 mitochondrial genomes. The genome distance shows high performance for quantitatively describing the differences between the gene organizations of the genomes. AVAILABILITY: Since the programs implementing these calculations require well-arranged gene organization data, they have not been released yet. However, one of the authors will analyze circular genomes upon request. Data on the gene organizations may be submitted electronically to the address below.

Two approximations were introduced into the double dynamic programming algorithm, in order to reduce the computational time for structural alignment. One of them was the so-called distance cut-off, which approximately describes the structural environment of each residue by its local environment. In the approximation, a sphere with a given radius is placed at the center of the side chain of each residue. The local environment of a residue is constituted only by the residues with side chain centers that are present within the sphere, which is expressed by a set of center-to-center distances from the side chain of the residue to those of all the other constituent residues. The residues outside the sphere are neglected from the local environment. Another approximation is associated with the distance cut-off, which is referred to here as the delta N cut-off. If two local environments are similar to each other, the numbers of residues constituting the environments are expected to be similar. The delta N cut-off was introduced based on the idea. If the difference between the numbers of the constituent residues of two local environments is greater than a given threshold value, delta N, the evaluation of the similarity between the local environments is skipped. The introduction of the two approximations dramatically reduced the computational time for structural alignment by the double dynamic programming algorithm. However, the approximations also decreased the accuracy of the alignment. To improve the accuracy with the approximations, a program with a two-step alignment algorithm was constructed. At first, an alignment was roughly constructed with the approximations. Then, the epsilon-suboptimal region for the alignment was determined. Finally, the double dynamic programming algorithm with full structural environments was applied to the residue pairs within the epsilon-suboptimal region to produce an improved alignment.