BACKGROUND: Regulation of gene expression, protein synthesis, replication and assembly of many viruses involve RNA-protein interactions. Although some successful computational tools have been reported to recognize RNA binding sites in proteins, the problem of specificity remains poorly investigated. After the nucleotide base composition, the dinucleotide is the smallest unit of RNA sequence information and many RNA-binding proteins simply bind to regions enriched in one dinucleotide. Interaction preferences of protein subsequences and dinucleotides can be inferred from protein-RNA complex structures, enabling a training-based prediction approach. RESULTS: We analyzed basic statistics of amino acid-dinucleotide contacts in protein-RNA complexes and found their pairing preferences could be identified. Using a standard approach to represent protein subsequences by their evolutionary profile, we trained neural networks to predict multiclass target vectors corresponding to 16 possible contacting dinucleotide subsequences. In the cross-validation experiments, the accuracies of the optimum network, measured as areas under the curve (AUC) of the receiver operating characteristic (ROC) graphs, were in the range of 65-80%. CONCLUSIONS: Dinucleotide-specific contact predictions have also been extended to the prediction of interacting protein and RNA fragment pairs, which shows the applicability of this method to predict targets of RNA-binding proteins. A web server predicting the 16-dimensional contact probability matrix directly from a user-defined protein sequence was implemented and made available at: http://tardis.nibio.go.jp/netasa/srcpred.

Genetic algorithm (GA) applied to feature selection and model optimization improved the performance of robust mathematical models such as Bayesian-regularized neural networks (BRANNs) and support vector machines (SVMs) on different drug design datasets. The selection of optimum input variables and the adjustment of network weights and biases to optimum values to yield generalizable predictors were optimized by combining Bayesian training and GA based-variable selection. Similarly, kernel and regularization parameters of SVMs were properly set by GA optimization. The predictors were more accurate and robust than previous published models on the same datasets. In addition, feature selection over large pools of molecular descriptors allowed determining the structural and atomic properties of the ligands that are ruling the biological interactions with the target.

BACKGROUND: The translation start site plays an important role in the control of translation efficiency of eukaryotic mRNAs. The recognition of the start AUG codon by eukaryotic ribosomes is considered to depend on its nucleotide context. However, the fraction of eukaryotic mRNAs with the start codon in a suboptimal context is relatively large. It may be expected that mRNA should possess some features providing efficient translation, including the proper recognition of a translation start site. It has been experimentally shown that a downstream hairpin located in certain positions with respect to start codon can compensate in part for the suboptimal AUG context and also increases translation from non-AUG initiation codons. Prediction of such a compensatory hairpin may be useful in the evaluation of eukaryotic mRNA translation properties. RESULTS: We evaluated interdependency between the start codon context and mRNA secondary structure at the CDS beginning: it was found that a suboptimal start codon context significantly correlated with higher base pairing probabilities at positions 13 - 17 of CDS of human and mouse mRNAs. It is likely that the downstream hairpins are used to enhance translation of some mammalian mRNAs in vivo. Thus, we have developed a tool, AUG_hairpin, to predict local stem-loop structures located within the defined region at the beginning of mRNA coding part. The implemented algorithm is based on the available published experimental data on the CDS-located stem-loop structures influencing the recognition of upstream start codons. CONCLUSION: An occurrence of a potential secondary structure downstream of start AUG codon in a suboptimal context (or downstream of a potential non-AUG start codon) may provide researchers with a testable assumption on the presence of additional regulatory signal influencing mRNA translation initiation rate and the start codon choice. AUG_hairpin, which has a convenient Web-interface with adjustable parameters, will make such an evaluation easy and efficient.

BACKGROUND: Detection of DNA-binding sites in proteins is of enormous interest for technologies targeting gene regulation and manipulation. We have previously shown that a residue and its sequence neighbor information can be used to predict DNA-binding candidates in a protein sequence. This sequence-based prediction method is applicable even if no sequence homology with a previously known DNA-binding protein is observed. Here we implement a neural network based algorithm to utilize evolutionary information of amino acid sequences in terms of their position specific scoring matrices (PSSMs) for a better prediction of DNA-binding sites. RESULTS: An average of sensitivity and specificity using PSSMs is up to 8.7% better than the prediction with sequence information only. Much smaller data sets could be used to generate PSSM with minimal loss of prediction accuracy. CONCLUSION: One problem in using PSSM-derived prediction is obtaining lengthy and time-consuming alignments against large sequence databases. In order to speed up the process of generating PSSMs, we tried to use different reference data sets (sequence space) against which a target protein is scanned for PSI-BLAST iterations. We find that a very small set of proteins can actually be used as such a reference data without losing much of the prediction value. This makes the process of generating PSSMs very rapid and even amenable to be used at a genome level. A web server has been developed to provide these predictions of DNA-binding sites for any new protein from its amino acid sequence. AVAILABILITY: Online predictions based on this method are available at http://www.netasa.org/dbs-pssm/

MOTIVATION: Though vitally important to cell function, the mechanism of protein-DNA binding has not yet been completely understood. We therefore analysed the relationship between DNA binding and protein sequence composition, solvent accessibility and secondary structure. Using non-redundant databases of transcription factors and protein-DNA complexes, neural network models were developed to utilize the information present in this relationship to predict DNA-binding proteins and their binding residues. RESULTS: Sequence composition was found to provide sufficient information to predict the probability of its binding to DNA with nearly 69% sensitivity at 64% accuracy for the considered proteins; sequence neighbourhood and solvent accessibility information were sufficient to make binding site predictions with 40% sensitivity at 79% accuracy. Detailed analysis of binding residues shows that some three- and five-residue segments frequently bind to DNA and that solvent accessibility plays a major role in binding. Although, binding behaviour was not associated with any particular secondary structure, there were interesting exceptions at the residue level. Over-representation of some residues in the binding sites was largely lost at the total sequence level, but a different kind of compositional preference was observed in DNA-binding proteins.

MOTIVATION: According to scanning model, 40S ribosomal subunits can either initiate translation at start AUG codon in suboptimal context or miss it and initiate translation at downstream AUG(s), thereby producing several proteins. Functional significance of such a protein translational polymorphism is still unknown. RESULTS: We compared predicted subcellular localizations of annotated Arabidopsis thaliana proteins and their potential N-terminally truncated forms started from the nearest downstream in-frame AUG codons. It was found that localizations of full and N-truncated proteins differ in many cases: 12.2% of N-truncated proteins acquired sorting signals de novo and 5.7% changed their predicted subcellular locations (mitochodria, chloroplast or secretory pathway). It is likely that the in-frame downstream AUGs may be frequently utilized to synthesize proteins possessing new functional properties and such a translational polymorphism may serve as an important source of cellular and organelle proteomes.

BACKGROUND: Accessible surface area (ASA) or solvent accessibility of amino acids in a protein has important implications. Knowledge of surface residues helps in locating potential candidates of active sites. Therefore, a method to quickly see the surface residues in a two dimensional model would help to immediately understand the population of amino acid residues on the surface and in the inner core of the proteins. RESULTS: ASAView is an algorithm, an application and a database of schematic representations of solvent accessibility of amino acid residues within proteins. A characteristic two-dimensional spiral plot of solvent accessibility provides a convenient graphical view of residues in terms of their exposed surface areas. In addition, sequential plots in the form of bar charts are also provided. Online plots of the proteins included in the entire Protein Data Bank (PDB), are provided for the entire protein as well as their chains separately. CONCLUSIONS: These graphical plots of solvent accessibility are likely to provide a quick view of the overall topological distribution of residues in proteins. Chain-wise computation of solvent accessibility is also provided.

SUMMARY: RVP-net is an online program for the prediction of real valued solvent accessibility. All previous methods of accessible surface area (ASA) predictions classify amino acid residues into exposure states and named them buried or exposed based on different thresholds. Real values in some cases were generated by taking the mid points of these state thresholds. This is the first method, which provides a direct prediction of ASA without making exposure categories and achieves results better than 19% mean absolute error. To facilitate batch processing of several sequences, a standalone version of this tool is also provided. AVAILABILITY: Online predictions are available at http://www.netasa.org/rvp-net/. Standalone version of the program can be obtained from the corresponding author by E-mail request.

MOTIVATION: Genome projects have produced large amounts of data on the sequences of new genes whose functions are as yet unknown. The functions of new genes are usually inferred by comparing their sequences with those of known genes, but evaluation of the sequence homology of individual genes does not make the most of the available sequence information. Therefore, new methods and tools for extracting more biological information from homology searches would be advantageous. RESULTS: We have developed a computational tool, ORI-GENE, to analyze the results of sequence homology searches from the perspective of the evolution of selected sets of new genes. ORI-GENE has a graphical interface and accomplishes two important tasks: first, based on the output of homology searches, it identifies species with similar genes and displays their pattern of distribution on the phylogenetic tree. This function enables one to infer the way in which a given gene may have propagated among species over time. Second, from the distribution patterns, it predicts the point at which a given gene may have been first acquired (i.e. its 'origin'), then classifies the gene on that basis. Because it makes use of available evolutionary information to show the way in which genes cluster among species, ORI-GENE should be an effective tool for the screening and classification of new genes revealed by genome analysis. AVAILABILITY: ORI-GENE is retrievable via the Internet at: http://www.rtc.riken.go.jp/jouhou/ORI-GENE.

MOTIVATION: Protein-nucleic acid interactions are fundamental to the regulation of gene expression. In order to elucidate the molecular mechanism of protein-nucleic acid recognition and analyze the gene regulation network, not only structural data but also quantitative binding data are necessary. Although there are structural databases for proteins and nucleic acids, there exists no database for their experimental binding data. Thus, we have developed a Thermodynamic Database for Protein-Nucleic Acid Interactions (ProNIT). RESULTS: We have collected experimentally observed binding data from the literature. ProNIT contains several important thermodynamic data for protein-nucleic acid binding, such as dissociation constant (K(d)), association constant (K(a)), Gibbs free energy change (DeltaG), enthalpy change (DeltaH), heat capacity change (DeltaC(p)), experimental conditions, structural information of proteins, nucleic acids and the complex, and literature information. These data are integrated into a relational database system together with structural and functional information to provide flexible searching facilities by using combinations of various terms and parameters. A www interface allows users to search for data based on various conditions, with different display and sorting options, and to visualize molecular structures and their interactions. AVAILABILITY: ProNIT is freely accessible at the URL http://www.rtc.riken.go.jp/jouhou/pronit/pronit.html.

MOTIVATION: Although a large amount of information on the structure, function and properties of biomolecules is becoming available, it is difficult to understand the relationship between them. Thus, we have attempted to create an integrated relational database, search and visualization tool, 3DinSight, to help researchers to gain insight into their relationship. RESULTS: We have gathered data on the structure, function and properties of biomolecules, and implemented them into a relational database system. The structural data contain several subset data such as protein homologues, protein-DNA complex, in order to enable searching within a specific class of data. The functional data include motif sequence and mutation data of proteins. Also, various amino acid properties are implemented as a relational table. The World Wide Web (WWW) interfaces enable users to carry out various kinds of searches among these data. The locations of motif sequences and mutations are automatically mapped on the structure, and visualized in three-dimensional (3D) space by interactive viewers, VRML (Virtual Reality Modeling Language) and RasMol. In the case of VRML, the mapped 3D objects are hyper-linked to the corresponding document data. Also, amino acid properties, linked with structure, functional and mutation sites, can be displayed as graph plots. AVAILABILITY: 3DinSight is freely accessible through the Internet (http://www.rtc.riken.go.jp/3DinSight.h tml). CONTACT: [email protected]