AGenDA: gene prediction by comparative sequence analysis

Comparative sequence analysis is a powerful approach to identify functional elements in genomic sequences. Herein, we describe AGenDA (Alignment-based GENe Detection Algorithm), a novel method for gene prediction that is based on long-range alignment of syntenic regions in eukaryotic genome sequences. Local sequence homologies identified by the DIALIGN program are searched for conserved splice signals to define potential protein-coding exons; these candidate exons are then used to assemble complete gene structures. The performance of our method was tested on a set of 105 human-mouse sequence pairs. These test runs showed that sensitivity and specificity of AGenDA are comparable with the best gene- prediction program that is currently available. However, since our method is based on a completely different type of input information, it can detect genes that are not detectable by standard methods and vice versa. Thus, our approach seems to be a useful addition to existing gene-prediction programs. Availability: DIALIGN is available through the Bielefeld Bioinformatics Server (BiBiServ) at http://bibiserv.techfak.uni-bielefeld.de/dialign/ The gene-prediction program AGenDA described in this paper will be available through the BiBiServ or MIPS web server at http://mips.gsf.de.     

A method of precise mRNA/DNA homology-based gene structure prediction

Background  

Accurate and automatic gene finding and structural prediction is a common problem in bioinformatics, and applications need to be capable of handling non-canonical splice sites, micro-exons and partial gene structure predictions that span across several genomic clones.     

ORB, a homology-based program for the prediction of protein NMR chemical shifts

A computer program (ORB) has been developed to predict ¹H,¹³C and ¹⁵N NMR chemical shifts of previouslyunassigned proteins. The program makes use of the information contained in achemical shift database of previously assigned proteins supplemented by astatistically derived averaged chemical shift database in which the shifts arecategorized according to their residue, atom and secondary structure type[Wishart et al. (1991) J. Mol. Biol., 222, 311–333]. The predictionprocess starts with a multiple alignment of all previously assigned proteinswith the unassigned query protein. ORB uses the sequence and secondarystructure alignment program XALIGN for this task [Wishart et al. (1994)CABIOS, 10, 121–132; 687–688]. The prediction algorithm in ORB isbased on a scoring of the known shifts for each sequence. The scores dependon global sequence similarity, local sequence similarity, structuralsimilarity and residue similarity and determine how much weight one particularshift is given in the prediction process. In situations where no applicablepreviously assigned chemical shifts are available, the shifts derived from theaveraged database are used. In addition to supplying the user with predictedchemical shifts, ORB calculates a confidence value for every prediction. Theseconfidence values enable the user to judge which predictions are the mostaccurate and they are particularly useful when ORB is incorporated into acomplete autoassignment package. The usefulness of ORB was tested on threemedium-sized proteins: an interleukin-8 analog, a troponin C synthetic peptideheterodimer and cardiac troponin C. Excellent results are obtained if ORB isable to use the chemical shifts of at least one highly homologous sequence.ORB performs well as long as the sequence identity between proteins with knownchemical shifts and the new sequence is not less than 30%.     

The discoidin domain family revisited: new members from prokaryotes and a homology-based fold prediction.

Members of the discoidin (DS) domain family, which includes the C1 and C2 repeats of blood coagulation factors V and VIII, occur in a great variety of eukaryotic proteins, most of which have been implicated in cell-adhesion or developmental processes. So far, no three-dimensional structure of a known example of this extracellular module has been determined, limiting the usefulness of identifying a new sequence as member of this family. Here, we present results of a recent search of the protein sequence database for new DS domains using generalized profiles, a sensitive multiple alignment-based search technique. Several previously unrecognized DS domains could be identified by this method, including the first examples from prokaryotic species. More importantly, we present statistical, structural, and functional evidence that the D1 domain of galactose oxidase whose three-dimensional structure has been determined at 1.7 A resolution, is a distant member of this family. Taken together, these findings significantly expand the concept of the DS domain, by extending its taxonomic range and by implying a fold prediction for all its members. The proposed alignment with the galactose oxidase sequence makes it possible to construct homology-based three-dimensional models for the most interesting examples, as illustrated by an accompanying paper on the C1 and C2 domains of factor V.     

Exploring dynamics of protein structure determination and homology-based prediction to estimate the number of superfamilies and folds

Background

As tertiary structure is currently available only for a fraction of known protein families, it is important to assess what parts of sequence space have been structurally characterized. We consider protein domains whose structure can be predicted by sequence similarity to proteins with solved structure and address the following questions. Do these domains represent an unbiased random sample of all sequence families? Do targets solved by structural genomic initiatives (SGI) provide such a sample? What are approximate total numbers of structure-based superfamilies and folds among soluble globular domains?

Results

To make these assessments, we combine two approaches: (i) sequence analysis and homology-based structure prediction for proteins from complete genomes; and (ii) monitoring dynamics of the assigned structure set in time, with the accumulation of experimentally solved structures. In the Clusters of Orthologous Groups (COG) database, we map the growing population of structurally characterized domain families onto the network of sequence-based connections between domains. This mapping reveals a systematic bias suggesting that target families for structure determination tend to be located in highly populated areas of sequence space. In contrast, the subset of domains whose structure is initially inferred by SGI is similar to a random sample from the whole population. To accommodate for the observed bias, we propose a new non-parametric approach to the estimation of the total numbers of structural superfamilies and folds, which does not rely on a specific model of the sampling process. Based on dynamics of robust distribution-based parameters in the growing set of structure predictions, we estimate the total numbers of superfamilies and folds among soluble globular proteins in the COG database.

Conclusion

The set of currently solved protein structures allows for structure prediction in approximately a third of sequence-based domain families. The choice of targets for structure determination is biased towards domains with many sequence-based homologs. The growing SGI output in the future should further contribute to the reduction of this bias. The total number of structural superfamilies and folds in the COG database are estimated as ~4000 and ~1700. These numbers are respectively four and three times higher than the numbers of superfamilies and folds that can currently be assigned to COG proteins.     

Protein function annotation by homology-based inference

With many genomes now sequenced, computational annotation methods to characterize genes and proteins from their sequence are increasingly important. The BioSapiens Network has developed tools to address all stages of this process, and here we review progress in the automated prediction of protein function based on protein sequence and structure.     

An automated homology-based approach for identifying transposable elements

Background  

Transposable elements (TEs) are mobile sequences found in nearly all eukaryotic genomes. They have the ability to move and replicate within a genome, often influencing genome evolution and gene expression. The identification of TEs is an important part of every genome project. The number of sequenced genomes is rapidly rising, and the need to identify TEs within them is also growing. The ability to do this automatically and effectively in a manner similar to the methods used for genes is of increasing importance. There exist many difficulties in identifying TEs, including their tendency to degrade over time and that many do not adhere to a conserved structure. In this work, we describe a homology-based approach for the automatic identification of high-quality consensus TEs, aimed for use in the analysis of newly sequenced genomes.     

GeneMachine: gene prediction and sequence annotation

MOTIVATION: A number of free-standing programs have been developed in order to help researchers find potential coding regions and deduce gene structure for long stretches of what is essentially 'anonymous DNA'. As these programs apply inherently different criteria to the question of what is and is not a coding region, multiple algorithms should be used in the course of positional cloning and positional candidate projects to assure that all potential coding regions within a previously-identified critical region are identified. RESULTS: We have developed a gene identification tool called GeneMachine which allows users to query multiple exon and gene prediction programs in an automated fashion. BLAST searches are also performed in order to see whether a previously-characterized coding region corresponds to a region in the query sequence. A suite of Perl programs and modules are used to run MZEF, GENSCAN, GRAIL 2, FGENES, RepeatMasker, Sputnik, and BLAST. The results of these runs are then parsed and written into ASN.1 format. Output files can be opened using NCBI Sequin, in essence using Sequin as both a workbench and as a graphical viewer. The main feature of GeneMachine is that the process is fully automated; the user is only required to launch GeneMachine and then open the resulting file with Sequin. Annotations can then be made to these results prior to submission to GenBank, thereby increasing the intrinsic value of these data. AVAILABILITY: GeneMachine is freely-available for download at http://genome.nhgri.nih.gov/genemachine. A public Web interface to the GeneMachine server for academic and not-for-profit users is available at http://genemachine.nhgri.nih.gov. The Web supplement to this paper may be found at http://genome.nhgri.nih.gov/genemachine/supplement/.     

In silico analysis of the tRNA:m1A58 methyltransferase family: homology-based fold prediction and identification of new members from Eubacteria and Archaea.

The amino acid sequences of Gcd10p and Gcd14p, the two subunits of the tRNA:(1-methyladenosine-58; m(1)A58) methyltransferase (MTase) of Saccharomyces cerevisiae, have been analyzed using iterative sequence database searches and fold recognition programs. The results suggest that the 'catalytic' Gcd14p and 'substrate binding' Gcd10p are related to each other and to a group of prokaryotic open reading frames, which were previously annotated as hypothetical protein isoaspartate MTases in sequence databases. It is predicted that the prokaryotic proteins are genuine tRNA:m(1)A MTases based on similarity of their predicted active site to the Gcd14p family. In addition to the MTase domain, an additional domain was identified in the N-terminus of all these proteins that may be involved in interaction with tRNA. These results suggest that the eukaryotic tRNA:m(1)A58 MTase is a product of gene duplication and divergent evolution of a possibly homodimeric prokaryotic enzyme.     

Thiol-dependent peroxidases care little about homology-based assignments of function

Abstract

Thiol-dependent peroxidase systems are reviewed with special emphasis on their potential use as drug targets. The basic catalytic mechanism of the two major thiol-peroxidase families, the glutathione peroxidases and the peroxiredoxins, are reasonably well understood. Sequence-based predictions of substrate specificities are still unsatisfactory. GPx-type enzymes are not generally specific for GSH but may specifically react with CXXC motifs as present in thioredoxins or tryparedoxins. Inversely, the peroxiredoxin family that was believed to be specific for CXXC-type proteins, also comprises glutathione peroxidases. Since structure-based predictions of function are also limited by small data bases, the increasing number of sequences emerging from genome projects require enzymatic characterization and genetic proof of relevance before they can be classified as drug targets.     

Thiol-dependent peroxidases care little about homology-based assignments of function

Thiol-dependent peroxidase systems are reviewed with special emphasis on their potential use as drug targets. The basic catalytic mechanism of the two major thiol-peroxidase families, the glutathione peroxidases and the peroxiredoxins, are reasonably well understood. Sequence-based predictions of substrate specificities are still unsatisfactory. GPx-type enzymes are not generally specific for GSH but may specifically react with CXXC motifs as present in thioredoxins or tryparedoxins. Inversely, the peroxiredoxin family that was believed to be specific for CXXC-type proteins, also comprises glutathione peroxidases. Since structure-based predictions of function are also limited by small data bases, the increasing number of sequences emerging from genome projects require enzymatic characterization and genetic proof of relevance before they can be classified as drug targets.     

Computational approaches to gene prediction

The problems associated with gene identification and the prediction of gene structure in DNA sequences have been the focus of increased attention over the past few years with the recent acquisition by large-scale sequencing projects of an immense amount of genome data. A variety of prediction programs have been developed in order to address these problems. This paper presents a review of the computational approaches and gene-finders used commonly for gene prediction in eukaryotic genomes. Two approaches, in general, have been adopted for this purpose: similarity-based and ab initio techniques. The information gleaned from these methods is then combined via a variety of algorithms, including Dynamic Programming (DP) or the Hidden Markov Model (HMM), and then used for gene prediction from the genomic sequences.     

Refinement of homology-based protein structures by molecular dynamics simulation techniques

The use of classical molecular dynamics simulations, performed in explicit water, for the refinement of structural models of proteins generated ab initio or based on homology has been investigated. The study involved a test set of 15 proteins that were previously used by Baker and coworkers to assess the efficiency of the ROSETTA method for ab initio protein structure prediction. For each protein, four models generated using the ROSETTA procedure were simulated for periods of between 5 and 400 nsec in explicit solvent, under identical conditions. In addition, the experimentally determined structure and the experimentally derived structure in which the side chains of all residues had been deleted and then regenerated using the WHATIF program were simulated and used as controls. A significant improvement in the deviation of the model structures from the experimentally determined structures was observed in several cases. In addition, it was found that in certain cases in which the experimental structure deviated rapidly from the initial structure in the simulations, indicating internal strain, the structures were more stable after regenerating the side-chain positions. Overall, the results indicate that molecular dynamics simulations on a tens to hundreds of nanoseconds time scale are useful for the refinement of homology or ab initio models of small to medium-size proteins.     

GeneBuilder: interactive in silico prediction of gene structure.

MOTIVATION: Prediction of gene structure in newly sequenced DNA becomes very important in large genome sequencing projects. This problem is complicated due to the exon-intron structure of eukaryotic genes and because gene expression is regulated by many different short nucleotide domains. In order to be able to analyse the full gene structure in different organisms, it is necessary to combine information about potential functional signals (promoter region, splice sites, start and stop codons, 3' untranslated region) together with the statistical properties of coding sequences (coding potential), information about homologous proteins, ESTs and repeated elements. RESULTS: We have developed the GeneBuilder system which is based on prediction of functional signals and coding regions by different approaches in combination with similarity searches in proteins and EST databases. The potential gene structure models are obtained by using a dynamic programming method. The program permits the use of several parameters for gene structure prediction and refinement. During gene model construction, selecting different exon homology levels with a protein sequence selected from a list of homologous proteins can improve the accuracy of the gene structure prediction. In the case of low homology, GeneBuilder is still able to predict the gene structure. The GeneBuilder system has been tested by using the standard set (Burset and Guigo, Genomics, 34, 353-367, 1996) and the performances are: 0.89 sensitivity and 0.91 specificity at the nucleotide level. The total correlation coefficient is 0.88. AVAILABILITY: The GeneBuilder system is implemented as a part of the WebGene a the URL: http://www.itba.mi. cnr.it/webgene and TRADAT (TRAncription Database and Analysis Tools) launcher URL: http://www.itba.mi.cnr.it/tradat.     

Combining gene prediction methods to improve metagenomic gene annotation

Background  

Traditional gene annotation methods rely on characteristics that may not be available in short reads generated from next generation technology, resulting in suboptimal performance for metagenomic (environmental) samples. Therefore, in recent years, new programs have been developed that optimize performance on short reads. In this work, we benchmark three metagenomic gene prediction programs and combine their predictions to improve metagenomic read gene annotation.     

Fast subcellular localization by cascaded fusion of signal-based and homology-based methods

Background

The functions of proteins are closely related to their subcellular locations. In the post-genomics era, the amount of gene and protein data grows exponentially, which necessitates the prediction of subcellular localization by computational means.

Results

This paper proposes mitigating the computation burden of alignment-based approaches to subcellular localization prediction by a cascaded fusion of cleavage site prediction and profile alignment. Specifically, the informative segments of protein sequences are identified by a cleavage site predictor using the information in their N-terminal shorting signals. Then, the sequences are truncated at the cleavage site positions, and the shortened sequences are passed to PSI-BLAST for computing their profiles. Subcellular localization are subsequently predicted by a profile-to-profile alignment support-vector-machine (SVM) classifier. To further reduce the training and recognition time of the classifier, the SVM classifier is replaced by a new kernel method based on the perturbational discriminant analysis (PDA).

Conclusions

Experimental results on a new dataset based on Swiss-Prot Release 57.5 show that the method can make use of the best property of signal- and homology-based approaches and can attain an accuracy comparable to that achieved by using full-length sequences. Analysis of profile-alignment score matrices suggest that both profile creation time and profile alignment time can be reduced without significant reduction in subcellular localization accuracy. It was found that PDA enjoys a short training time as compared to the conventional SVM. We advocate that the method will be important for biologists to conduct large-scale protein annotation or for bioinformaticians to perform preliminary investigations on new algorithms that involve pairwise alignments.