Exon Prediction Research Articles

Dynamic visual data mining: biological sequence analysis and annotation using SeqVISTA

In the post-genomic era, the volume of public sequence databases is increasing exponentially and visualisation-centric techniques have become more and more important in biological sequence analysis and annotation. In this paper, we present a methodology called dynamic visual data mining (DVDM), which combines biological object modelling, interactive display, and data analysis tools into one integrative platform. Using Java Development Kit v1.4, an object-oriented software named SeqVISTA has been developed based on DVDM. To illustrate the application of SeqVISTA, the following examples are shown: regular expression pattern matching; comparative analysis of alternative exon splicing patterns; Fourier analyses; exon prediction (MZEF and GENSCAN). Overall, we argue that DVDM is an important technique for biologists to unveil the information hidden behind the large genomic and proteomic databases, and SeqVISTA provides a versatile tool that integrates multiple computational algorithms for meeting biologists' data mining needs.

Chaining algorithms for multiple genome comparison

Given n fragments from k > 2 genomes, Myers and Miller showed how to find an optimal global chain of colinear non-overlapping fragments in O ( n log k n ) time and O ( n log k − 1 n ) space. For gap costs in the L 1 -metric, we reduce the time complexity of their algorithm by a factor log 2 n log log n and the space complexity by a factor log n . For the sum-of-pairs gap cost, our algorithm improves the time complexity of their algorithm by a factor log n log log n . A variant of our algorithm finds all significant local chains of colinear non-overlapping fragments. These chaining algorithms can be used in a variety of problems in comparative genomics: the computation of global alignments of complete genomes, the identification of regions of similarity (candidate regions of conserved synteny), the detection of genome rearrangements, and exon prediction.

Comparative genomics: methods and applications.

Interpreting the functional content of a given genomic sequence is one of the central challenges of biology today. Perhaps the most promising approach to this problem is based on the comparative method of classic biology in the modern guise of sequence comparison. For instance, protein-coding regions tend to be conserved between species. Hence, a simple method for distinguishing a functional exon from the chance absence of stop codons is to investigate its homologue from closely related species. Predicting regulatory elements is even more difficult than exon prediction, but again, comparisons pinpointing conserved sequence motifs upstream of translation start sites are helping to unravel gene regulatory networks. In addition to interspecific studies, intraspecific sequence comparison yields insights into the evolutionary forces that have acted on a species in the past. Of particular interest here is the identification of selection events such as selective sweeps. Both intra- and interspecific sequence comparisons are based on a variety of computational methods, including alignment, phylogenetic reconstruction, and coalescent theory. This article surveys the biology and the central computational ideas applied in recent comparative genomics projects. We argue that the most fruitful method of understanding the functional content of genomes is to study them in the context of related genomic sequences. In particular, such a study may reveal selection, a fundamental pointer to biological relevance.

From PREDs and Open Reading Frames to cDNA Isolation: Revisiting the Human Chromosome 21 Transcription Map

A supernumerary copy of human chromosome 21 (HC21) causes Down syndrome. To understand the molecular pathogenesis of Down syndrome, it is necessary to identify all HC21 genes. The first annotation of the sequence of 21q confirmed 127 genes, and predicted an additional 98 previously unknown “anonymous” genes (predictions (PREDs) and open reading frames (C21orfs)), which were foreseen by exon prediction programs and/or spliced expressed sequence tags. These putative gene models still need to be confirmed as bona fide transcripts. Here we report the characterization and expression pattern of the putative transcripts C21orf7, C21orf11, C21orf15, C21orf18, C21orf19, C21orf22, C21orf42, C21orf50, C21orf51, C21orf57, and C21orf58, the GC-rich sequence DNA-binding factor candidate GCFC (also known as C21orf66), PRED12, PRED31, PRED34, PRED44, PRED54, and PRED56. Our analysis showed that most of the C21orfs originally defined by matching spliced expressed sequence tags were correctly predicted, whereas many of the PREDs, defined solely by computer prediction, do not correspond to genuine genes. Four of the six PREDs were incorrectly predicted: PRED44 and C21orf11 are portions of the same transcript, PRED31 is a pseudogene, and PRED54 and PRED56 were wrongly predicted. In contrast, PRED12 (now called C21orf68) and PRED34 (C21orf63) are now confirmed transcripts. We identified three new genes, C21orf67, C21orf69, and C21orf70, not previously predicted by any programs. This revision of the HC21 transcriptome has consequences for the entire genome regarding the quality of previous annotations and the total number of transcripts. It also provides new candidates for genes involved in Down syndrome and other genetic disorders that map to HC21.

Computational analysis and prediction for exons of PAC579 genomic sequence

To isolate the novel genes related to human hepatocellular carcinoma (HCC), we sequenced P1-derived artificial chromosome PAC579 (D17S926 locus) mapped in the minimum LOH (loss of heterozygosity) deletion region of chromosome 17p13.3 in HCC. Four novel genes mapped in this genomic sequence area were isolated and cloned by wet-lab experiments, and the exons of these genes were located. 0-60 kb of this genomic sequence including the genes of interest was scanned with five different computational exon prediction programs as well as four splice site recognition programs. After analyzing and comparing the computationally predicted results with the wet-lab experiment results, some potential exons were predicted in the genomic sequence by using these programs.

Prediction of many new exons and introns in Plasmodium falciparum chromosome 2

The current prediction of genes in the Plasmodium falciparum genome database relies upon a limited number of specially developed computer algorithms. We have re-annotated the sequence of chromosome 2 of P. falciparum by a computer-assisted manual analysis, which is described here. Of 161 newly predicted introns, we have experimentally confirmed 98. We regard 110 introns from the previously published analyses as probable, we delete 3, change 26 and add 135. We recognise 214 genes in chromosome 2. We have predicted introns in 121 genes. The increased complexity of gene structure on chromosome 2 is likely to be mirrored by the entire genome.

Identification of a 350-kDa ClpP Protease Complex with 10 Different Clp Isoforms in Chloroplasts of Arabidopsis thaliana

A 350-kDa ClpP protease complex with 10 different subunits was identified in chloroplast of Arabidopsis thaliana, using Blue-Native gel electrophoresis, followed by matrix-assisted laser desorption ionization time-of-flight and nano-electrospray tandem mass spectrometry. The complex was copurified with the thylakoid membranes, and all identified Clp subunits show chloroplast targeting signals, supporting that this complex is indeed localized in the chloroplast. The complex contains chloroplast-encoded pClpP and six nuclear-encoded proteins nCpP1-6, as well as two unassigned Clp homologues (nClpP7, nClpP8). An additional Clp protein was identified in this complex; it does not belong to any of the known Clp genes families and is here assigned ClpS1. Expression and accumulation of several of these Clp proteins have never been shown earlier. Sequence and phylogenetic tree analysis suggests that nClpP5, nClpP2, and nClpP8 are not catalytically active and form a new group of Clp higher plant proteins, orthologous to the cyanobacterial ClpR protein, and are renamed ClpR1, -2, and -3, respectively. We speculate that ClpR1, -2, and -3 are part of the heptameric rings, whereas ClpS1 is a regulatory subunit positioned at the axial opening of the ClpP/R core. Several truncations and errors in intron and exon prediction of the annotated Clp genes were corrected using mass spectrometry data and by matching genomic sequences with cDNA sequences. This strategy will be widely applicable for the much needed verification of protein prediction from genomic sequence. The extreme complexity of the chloroplast Clp complex is discussed.

Toward Chromosome-Level Comparative Analysis of Vertebrate Genomes

Since the publication of the human draft genome sequence, steady progress around the world is made to analyze the genome. In addition, mouse genome project and other vertebrate genome projects of model organisms are rapidly producing much genomic sequences. Comparative analysis is becoming increasingly important as a tool for genome sequence analysis [3]. In this study, we focus on vertebrate genome sequences of various species, such as human, mouse and rat. The evolutionary conserved regions in human genome sequences are identified effectively by the comparison with those vertebrate genome sequences. The main goal of this research is to identify novel genes and regulatory regions and to elucidate genome-level events in evolution from genome sequences. We have developed a comparative genome system which collects data, compares them, visualizes the results, and produces primer sets using primer3 [6] for the PCR experiments for confirmation of the exon prediction. Users can submit syntenic genomic sequences or select a publicly available syntenic region. The system allow the users to analyze and convert data easily, so that users can browse the various analysis results using PipMaker [4] (WWW server for genome comparison and java application laj), and Alfresco [2] (java application), without producing annotation files on their own.

MAGPIE/EGRET annotation of the 2.9-Mb Drosophila melanogaster Adh region.

Our challenge in annotating the 2.91-Mb Adh region of the Drosophila melanogaster genome was to identify genetic and genomic features automatically, completely, and precisely within a 6-week period. To do so, we augmented the MAGPIE microbial genome annotation system to handle eukaryotic genomic sequence data. The new configuration required the integration of eukaryotic gene-finding tools and DNA repeat tools into the automatic data collection module. It also required us to define in MAGPIE new strategies to combine data about eukaryotic exon predictions with functional data to refine the exon predictions. At the heart of the resulting new eukaryotic genome annotation system is a reverse comparison of public protein and complementary DNA sequences against the input genome to identify missing exons and to refine exon boundaries. The software modules that add eukaryotic genome annotation capability to MAGPIE are available as EGRET (Eukaryotic Genome Rapid Evaluation Tool).

Mapping of the POF1 locus and identification of putative genes for premature ovarian failure.

We have identified a breakpoint on the X chromosome which is associated with premature ovarian failure (POF). Using polymerase chain reaction (PCR) probes of polymorphic microsatellites and fluorescent in-situ hybridization (FISH), this breakpoint has been narrowed to a region of 300 kb spanned by two P1 artificial chromosomes (PAC). Computer exon prediction and gene homology programs revealed three genes in this area. Our results suggest that two of these genes, HS6ST and E2F, and LINE 1 elements may be involved in ovarian development. Interruption of these genes could be the cause of POF. This study demonstrates how various molecular techniques and bioinformatic searches can complement each other in order to solve a clinical problem.

CDNA cloning and genomic structure of a novel gene (C11orf9) localized to chromosome 11q12→q13.1 which encodes a highly conserved, potential membrane-associated protein

We have cloned and characterized a novel gene (C11orf9) mapping to chromosome 11q12→q13.1. The transcript was initially identified as a partial cDNA sequence in the course of constructing a transcript map of the region between markers D11S1765 and uteroglobin known to encompass the gene causing Best disease. Using a combination of EST mapping, computational exon prediction, RT-PCR, and 5′-RACE its 5.7-kb full-length cDNA sequence was subsequently obtained. The C11orf9 gene consists of 26 exons spanning 33.1 kb of genomic DNA and is located about 4.3 kb centromeric to FEN1. Biocomputational analysis predicts that its conceptualtranslation product of 1,111 amino acids contains two transmembrane helices as well as two proline-rich regions. Alignment reveals significant homology to hypothetical peptides from several other species including C. elegans and D. melanogaster, indicating a high degree of conservation throughout evolution. Northern Blot and RT-PCR analyses demonstrate widespread expression of a single transcript but varying degrees of abundance among the individual tissues tested. Mutation analysis of the entire coding sequence excluded C11orf9 as the Best disease gene.

A Dictionary-Based Approach for Gene Annotation

This paper describes a fast and fully automated dictionary-based approach to gene annotation and exon prediction. Two dictionaries are constructed, one from the nonredundant protein OWL database and the other from the dbEST database. These dictionaries are used to obtain O (1) time lookups of tuples in the dictionaries (4 tuples for the OWL database and 11 tuples for the dbEST database). These tuples can be used to rapidly find the longest matches at every position in an input sequence to the database sequences. Such matches provide very useful information pertaining to locating common segments between exons, alternative splice sites, and frequency data of long tuples for statistical purposes. These dictionaries also provide the basis for both homology determination, and statistical approaches to exon prediction.

The FSHD Region on Human Chromosome 4q35 Contains Potential Coding Regions among Pseudogenes and a High Density of Repeat Elements

The distal end of chromosome 4q contains the locus involved in facioscapulohumeral muscular dystrophy (FSHD1). Specific genomic deletions within a tandem DNA repeat (D4Z4) are associated with the disease status, but no causal genes have yet been discovered. In a systematic search for genes, a 161-kb stretch of genomic DNA proximal to D4Z4 was sequenced, analyzed for homologies, and subjected to gene prediction programs. A major fraction (45%) of the subtelomeric region is composed of repeat sequences attributable mainly to LINE-1 elements. Apart from the previously identified FRG1 and TUB4q sequences, several additional potential coding regions were identified by analyzing the sequence with exon prediction programs. So far, we have been unable to demonstrate transcripts by RT-PCR or cDNA library hybridization. However, several retrotransposed pseudogenes were identified. The high density of pseudogenes and repeat elements is consistent with the subtelomeric location of this region and explains why previous transcript identification studies have been problematic.

Comparative Sequence Analysis of the Mouse and Human Lgn1/SMA Interval

Human chromosome 5q11.2–q13.3 and its ortholog on mouse chromosome 13 contain candidate genes for an inherited human neurodegenerative disorder called spinal muscular atrophy (SMA) and for an inherited mouse susceptibility to infection with Legionella pneumophila (Lgn1). These homologous genomic regions also have unusual repetitive organizations that create practical difficulties in mapping and raise interesting issues about the evolutionary origin of the repeats. In an attempt to analyze this region in detail, and as a way to identify additional candidate genes for these diseases, we have determined the sequence of 179 kb of the mouse Lgn1/SMA interval. We have analyzed this sequence using BLAST searches and various exon prediction programs to identify potential genes. Since these methods can generate false-positive exon declarations, our alignments of the mouse sequence with available human orthologous sequence allowed us to discriminate rapidly among this collection of potential coding regions by indicating which regions were well conserved and were more likely to represent actual coding sequence. As a result of our analysis, we accurately mapped two additional genes in the SMA interval that can be tested for involvement in the pathogenesis of SMA. While no new Lgn1 candidates emerged, we have identified new genetic markers that exclude Smn as an Lgn1 candidate. In addition to providing important resources for studying SMA and Lgn1, our data provide further evidence of the value of sequencing the mouse genome as a means to help with the annotation of the human genomic sequence and vice versa.

Identification and Characterization of the Human Homologue (RAI2) of a Mouse Retinoic Acid-Induced Gene in Xp22

We have identified a novel human gene during studies of a 1.3-Mb region of Xp22 between DXS418 and DXS999. A PAC contig spanning the region was constructed, sequenced, and analyzed by gene and exon prediction programs and by homology searches. Further investigation of predicted exons from PAC clone 389A20 led to the identification of a single-exon gene, designatedRAI2(retinoic acid-induced 2).RAI2mapped 28 kb centromeric to marker DXS7996, between DXS7996 and DXS7997, and was transcribed from centromere to telomere. Northern blot analysis and reverse transcription-polymerase chain reaction analysis revealed expression of a 2.5-kb transcript in four fetal tissues (brain, lung, kidney, and heart) and eight adult tissues (heart, brain, placenta, lung, skeletal muscle, kidney, pancreas, and retina) but not in fetal or adult liver. The 530-amino-acid protein (57 kDa predicted mass) displays 94% homology with a mouse retinoic acid-induced gene product and contains a novel proline-rich (39%) domain of 68 amino acids. Retinoic acid is involved in vertebrate anteroposterior axis formation and cellular differentiation and has been shown to modulate gene expression controlling early embryonal development, suggesting a developmental role forRAI2. RAI2remains a candidate gene for diseases mapping to the Xp22 region.

5] Current status of computational gene finding: A perspective

This chapter discusses the present status of computational approaches to gene finding and its utility. It must be stressed that a gene predicted by computational methods must be viewed as a hypothesis subject to experimental verification. In some cases, there may be strong evidence to support the model, the transcript of the model gene is an exact match to the sequence of a cDNA in GenBank, for example, but in many cases, confirmation of the prediction will come only from the generation of further data. Another distinction that needs to be made is between finding a gene and providing a complete and accurate description of it. Often, the correct prediction of a single exon is sufficient to identify, and properly locate, a gene in a genomic sequence even if its complete intron-exon structure is not properly represented. The chapter discusses some of the limitations of the present approaches and their causes.

6] Gene identification by exon amplification

For humans as well as other organisms with more complex genomes, development of maps provide a powerful tool in positional cloning efforts, but this is currently a daunting task. Whole-genome, chromosome-specific, and region-specific endeavors employ a variety of approaches that focus on characteristics that are common to most genes. Although sequence-based exon prediction methods can be useful for identifying genes and determining gene structure, it is currently unclear whether computer-based gene prediction algorithms will be reliable for comprehensive identification of genes in genomes of this complexity. Another set of methods relies on the biological selection of functional motifs in genomic DNA. This chapter discusses exon amplification, which is the most widely used of the functional selection methods for gene identification.

Large-scale sequencing of two regions in human chromosome 7q22: analysis of 650 kb of genomic sequence around the EPO and CUTL1 loci reveals 17 genes.

We have sequenced and annotated two genomic regions located in the Giemsa negative band q22 of human chromosome 7. The first region defined by the erythropoietin (EPO) locus is 228 kb in length and contains 13 genes. Whereas 3 genes (GNB2, EPO, PCOLCE) were known previously on the mRNA level, we have been able to identify 10 novel genes using a newly developed automatic annotation tool RUMMAGE-DP, which comprises >26 different programs mainly for exon prediction, homology searches, and compositional and repeat analysis. For precise annotation we have also resequenced ESTs identified to the region and assembled them to build large cDNAs. In addition, we have investigated the differential splicing of genes. Using these tools we annotated 4 of the 10 genes as a zonadhesin, a transferrin homolog, a nucleoporin-like gene, and an actin gene. Two genes showed weak similarity to an insulin-like receptor and a neuronal protein with a leucine-rich amino-terminal domain. Four predicted genes (CDS1-CDS4) CDS that have been confirmed on the mRNA level showed no similarity to known proteins and a potential function could not be assigned. The second region in 7q22 defined by the CUTL1 (CCAAT displacement protein and its splice variant) locus is 416 kb in length and contains three known genes, including PMSL12, APS, CUTL1, and a novel gene (CDS5). The CUTL1 locus, consisting of two splice variants (CDP and CASP), occupies >300 kb. Based on the G, C profile an isochore switch can be defined between the CUTL1 gene and the APS and PMSL12 genes.

Pombe: a gene-finding and exon-intron structure prediction system for fission yeast.

A special program developed by the authors, called Pombe, identifies protein coding regions in the Schizosaccharomyces pombe genome. Linear discriminant analysis was applied to predict 5'-terminal, internal, 3'-terminal exons (coding-exon) and introns. The accuracy of the prediction was tested by cross verifications. The sensitivity, specificity and correlation coefficient for the internal exon prediction were 98.5%, 99.9% and 98.3% respectively at the nucleotide level. Open reading frames were studied and used to predict intron-less genes: 99.0% of such genes were identified with correct stopping sites. The gene structure was determined by dynamic programming and the prediction achieved 97.0% correlation coefficient at the nucleotide level. The program is available at http:(/)/clio.cshl.org/genefinder.

Pombe: A gene-finding and exon-intron structure prediction system for fission yeast

A special program developed by the authors, called Pombe, identifies protein coding regions in the Schizosaccharomyces pombe genome. Linear discriminant analysis was applied to predict 5'-terminal, internal, 3'-terminal exons (coding-exon) and introns. The accuracy of the prediction was tested by cross verifications. The sensitivity, specificity and correlation coefficient for the internal exon prediction were 98.5%, 99.9% and 98.3% respectively at the nucleotide level. Open reading frames were studied and used to predict intron-less genes: 99.0% of such genes were identified with correct stopping sites. The gene structure was determined by dynamic programming and the prediction achieved 97.0% correlation coefficient at the nucleotide level. The program is available at http:(/)/clio.cshl.org/genefinder.