Dataset Construction for Gene Structure Prediction and Alternative Splicing Analysis

The Drosophila gene Dstpk61 encodes a serine threonine protein kinase homologous to human phosphoinositide-dependent protein kinase (PDK1), and also has homologues in S. cerevisiae, S. pombe, C. elegans, A. thaliana, mouse, and sheep. Where its function has been investigated, this kinase is thought to be involved in regulating cell growth and survival in response to extracellular signals such as insulin and growth factors. In Drosophila it produces multiple transcripts, some of which appear to be sex-specific. In addition to the five Dstpk61 cDNAs we have described previously we report the existence of a further 18 expressed sequence tag (EST) cDNAs, three of which we have fully sequenced. We conclude that Dstpk61 is a complex locus that utilises a combination of alternative promoters, alternative splice sites and alternative polyadenylation sites to produce a vast array of different transcripts. These cDNAs encode at least four different DSTPK61 protein isoforms with variant N-termini. In this paper, we discuss the possible functions of the distinct Dstpk61 transcripts and how they might be differentially regulated. We also discuss the roles that DSTPK61 protein isoforms might play in relation to the protein domains they contain and their potential targets in the cell. Finally, we report the putative structure of the human PDK1 gene based on computer comparisons of available mRNA and genomic sequences. The value of using sequence data from other species for experimental design in mammalian systems is discussed.

Its accessibility, unique evolutionary position, and recently assembled genome sequence have advanced the chicken to the forefront of comparative genomics and developmental biology research as a model organism. Several chicken expressed sequence tag (EST) projects have placed the chicken in 10th place for accrued ESTs among all organisms in GenBank. We have completed the single-pass 5'-end sequencing of 37,557 chicken cDNA clones from several single and multiple tissue cDNA libraries and have entered 35,407 EST sequences into GenBank. Our chicken EST sequences and those found in public databases (on July 1, 2004) provided a total of 517,727 public chicken ESTs and mRNAs. These sequences were used in the CAP3 assembly of a chicken gene index composed of 40,850 contigs and 79,192 unassembled singlets. The CAP3 contigs show a 96.7% match to the chicken genome sequence. The University of Delaware (UD) EST collection (43,928 clones) was assembled into 19,237 nonredundant sequences (13,495 contigs and 5,742 unassembled singlets). The UD collection contains 6,223 unique sequences that are not found in other public EST collections but show a 76% match to the chicken genome sequence. Our chicken contig and singlet sequences were annotated according to the highest BlastX and/or BlastN hits. The UD CAP3 contig assemblies and singlets are searchable by nucleotide sequence or key word (http://cogburn.dbi.udel.edu), and the cDNA clones are readily available for distribution from the chick EST website and clone repository (http://www.chickest.udel.edu). The present paper describes the construction and normalization of single and multiple tissue chicken cDNA libraries, high-throughput EST sequencing from these libraries, the CAP3 assembly of a chicken gene index from all public ESTs, and the identification of several nonredundant chicken gene sets for production of custom DNA microarrays.

To investigate the gene structures of RHD mRNA alternative splicing isoforms, total RNA was extracted from normal cord blood samples and RHD mRNA were detected using reverse transcription and polymerase chain reaction (RT-PCR) method. The PCR products were cloned and directly sequenced. The splicing signals of various isoforms were also analyzed by sequencing of exon-intron boundaries using RHD gene specific primers, and the expressed sequence tags (ESTs) databank was screened to find out other RHD alternative splicing isoforms. Nine novel and three recurrent RHD alternative splicing isoforms were identified, among which six isoforms were RHD-RHCE hybrid genes. Three types of alternative splicing were found including exon skipping, alternative 5' splice sites and alternative 3' splice sites. Exons 2 to 9 were involved in RHD mRNA alternative splicing. Two other isoforms of intron retention type were also found in human ESTs databank. Thus, the mRNA splicing mechanism of RHD gene was very complex, and nine novel alternative splicing isoforms were identified including several isoforms based on homologous RHD-RHCE hybrid gene.

There is a pressing need to align growing set of expressed sequence tags (ESTs) to newly sequenced human genome that is still frequently revised, for providing biologists and medical scientists with fresh information. The problem is, however, complicated by the exon/intron structure of eucaryotic genes, misread nucleotides in ESTs, and millions of repeptive sequences in genomic sequences. Indeed, to solve this, algorithms that use dynamic programming have been proposed, in which space complexity is O(N) and time complexity is O(MN) for a genomic sequence of length M and an EST of length N, but in reality, these algorithms require an enormous amount of processing time. In an effort to improve the computational efficiency of these classical DP algorithms, we develop software that fully utilizes the lookup-table that stores the position at which each short subsequence occurs in the genomic sequence for allowing the efficient detection of the start-and endpoints of an EST within a given DNA sequence, and subsequently, the prompt identification of exons and introns. In addition, high sensitivity and accuracy must be achieved by calculating locations of all spliced sites correctly for more ESTs while retaining high computational efficiency. This goal is hard to accomplish in practice, owing to misread nucleotides in ESTs and repeptive sequences in the genome, but we present a couple of heuristics effective in settling this issue. Experimental results have confirmed that our technique improves the overall computation time by orders of magnitude compared with common tools such as sim4 and BLAT, and attains high sensitivity and accuracy against datasets of clean and documented genes at the same time. Consequently, our software is able to align about three millions of ESTs to a draft genome in less than one day, and all the information is available through the WWW at http://grl.gi.k.u-tokyo.ac.jp/.KeywordsHuman GenomeDynamic ProgrammingMatch RatioMultiple ExonPractical SoftwareThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

In the past several years, the analysis of alternative splicing and its application to identify the origins of disease has been gaining momentum. Therefore, we developed a value added transcriptome database (Avatar). We mapped EST (expressed sequence tag) and mRNA sequences onto whole human genomic sequence. We identified 9,937 alternative splicing relationships genes through an analysis of large-scale ESTs. Besides, we provided alternative splicing information for six organisms: Homo sapiens, Mus musculus, Rattus norvegiens, Caenorhabditis elegans, Drosophila melanogaster and Arabidopsis thaliana.

Several reports addressed the issue of how the alternative splicing of exon 7 and 8 in the APP pre-mRNA is regulated in different tissues. Of special interest here was the potential involvement of exon 7 containing APP splice isoforms, since this exon codes for a serine protease inhibitor and is therefore of putative relevance for amyloidogenic catabolism of the precursor protein. The recent identification of a third alternative splice site in close proximity to the beta A4-amyloid portion in the APP gene which may also increase APP amyloidogenicity, allowed us to investigate its regulation in cells of the central nervous system. With our assay, we were able to resolve six different APP isoforms of the eight potential isoforms which can be generated from the three alternatively spliced exons 7, 8, and 15. We demonstrate here that, in addition to rat brain microglia cells, astrocyte-enriched cultures also skip the novel alternative 3'-splice site in front of exon 15, generating L-APP mRNA. Neurons are the only cells in the central nervous system which seem to use the 3'-splice site of intron 14 nearly 100%. Interestingly, this very 3'-splice site is the only one present in the APP gene that completely matches the consensus sequence for the branchpoint sequence proposed for introns. We would therefore suggest that neurons lack a specific splicing factor which inhibits the use of the rather strong 3'-splice site in front of exon 15. It remains to be shown whether this is also the case for neurons in Alzheimer's disease.

Single-nucleotide polymorphisms (SNPs) are desirable DNA markers for marker-assisted selection because of their high abundance, codominant inheritance, locus specificity and potential for automated high-throughput analysis. The average number of SNPs in introns has been reported to be several times higher than that in exons in many organisms. Potential intron polymorphism (PIP) markers, designed from expressed sequence tag (EST) sequences (exon regions), could amplify potential flanking intron regions without requiring genome sequence information. In this study, we used PIP markers designed from apple (Malus × domestica) ESTs to identify intron regions and to detect SNPs in pear (Pyrus spp.). We evaluated 170 PIP markers for Japanese pear P. pyrifolia cultivar 'Housui' and the European pear P. communis cultivar 'Bartlett'. Of the 170 PIP markers, 142 markers amplified fragments in both cultivars. The fragment size was larger than the size predicted from the apple EST sequences, suggesting the existence of introns in the amplified fragments. We sequenced the amplified fragments and detected candidate bi-allelic SNPs in intron regions of both cultivars. The genome sequences obtained from pear were aligned to the apple EST sequences, revealing that the sequences included donor and acceptor sites, and contained deduced exons and flanking introns in pear. SNP analysis showed that 55 and 54 markers were mapped on the genetic linkage maps of 'Bartlett' and 'Housui', respectively. The PIP markers designed from apple ESTs were applicable in pear and could be used to detect SNPs in intron regions and to develop SNP markers in pear. The PIP markers will facilitate marker-assisted selection for disease resistance and other important traits in pear breeding programs.

SSR markers have become the most popular resource for studying population genetic variation in eukaryotes. However, few studies with SSR markers have been carried out with phytoparasitic nematodes. In this study a primary survey on EST-SSRs was made utilizing bioinformatics methods to derive SSRs from expressed sequence tags (ESTs) of 16 species of PPN, which belong to 7 genera and 5 families. The results showed that trinucleotide repeats were the most abundant SSRs in coding ESTs, while tetranucleotide SSRs were predominant in non-coding ESTs and genome sequences. AG/CT, AAC/GTT and (AAAN)n motifs were predominant, and CG/GC, ACT/AGT motifs were scanty in the ESTs of most genera and species. SSRs were more abundant in non-coding ESTs than in coding ESTs. The distribution frequencies of SSR motifs in coding ESTs, non-coding ESTs and genomes are different. Our results will provide useful information for screening SSRs from each genus and species for further genetic study of phytoparastic nematodes.

BackgroundUntil recently, few genomic reagents specific for non-human primate research have been available. To address this need, we have constructed a macaque-specific high-density oligonucleotide microarray by using highly fragmented low-pass sequence contigs from the rhesus genome project together with the detailed sequence and exon structure of the human genome. Using this method, we designed oligonucleotide probes to over 17,000 distinct rhesus/human gene orthologs and increased by four-fold the number of available genes relative to our first-generation expressed sequence tag (EST)-derived array.ResultsWe constructed a database containing 248,000 exon sequences from 23,000 human RefSeq genes and compared each human exon with its best matching sequence in the January 2005 version of the rhesus genome project list of 486,000 DNA contigs. Best matching rhesus exon sequences for each of the 23,000 human genes were then concatenated in the proper order and orientation to produce a rhesus "virtual transcriptome." Microarray probes were designed, one per gene, to the region closest to the 3' untranslated region (UTR) of each rhesus virtual transcript. Each probe was compared to a composite rhesus/human transcript database to test for cross-hybridization potential yielding a final probe set representing 18,296 rhesus/human gene orthologs, including transcript variants, and over 17,000 distinct genes. We hybridized mRNA from rhesus brain and spleen to both the EST- and genome-derived microarrays. Besides four-fold greater gene coverage, the genome-derived array also showed greater mean signal intensities for genes present on both arrays. Genome-derived probes showed 99.4% identity when compared to 4,767 rhesus GenBank sequence tag site (STS) sequences indicating that early stage low-pass versions of complex genomes are of sufficient quality to yield valuable functional genomic information when combined with finished genome information from a closely related species.ConclusionThe number of different genes represented on microarrays for unfinished genomes can be greatly increased by matching known gene transcript annotations from a closely related species with sequence data from the unfinished genome. Signal intensity on both EST- and genome-derived arrays was highly correlated with probe distance from the 3' UTR, information often missing from ESTs yet present in early-stage genome projects.

The relationship between polyadenylation and splicing was investigated in a model system consisting of two tandem but nonidentical polyomavirus late transcription units. This model system exploits the polyomavirus late transcription termination and polyadenylation signals, which are sufficiently weak to allow the production of many multigenome-length primary transcripts with repeating introns, exons, and poly(A) sites. This double-genome construct contains exons of two types, those bordered by 3' and 5' splice sites (L1 and L2) and those bordered by a 3' splice site and a poly(A) site (V1 and V2). The L1 and L2 exons are distinguishable from one another but retain identical flanking RNA processing signals, as is the case for the V1 and V2 exons. Analysis of cytoplasmic RNAs obtained from mouse cells transfected with this construct and its derivatives revealed the following. (i) V1 and V2 exons are often skipped during pre-mRNA processing, while L1 and L2 exons are not skipped. (ii) No messages contain internal, unused polyadenylation signals. (iii) Poly(A) site choice is not required for the selection of an upstream 3' splice site. (iv) When two tandem poly(A) sites are placed downstream of a 3' splice site, the first poly(A) site is chosen almost exclusively, even though transcription can proceed past both sites. (v) Placing a 3' splice site between these two tandem poly(A) sites allows the more distal site to be chosen. These and other available data are most consistent with a model in which terminal exons are produced by the coordinate selection and use of a 3' splice site with the nearest available downstream poly(A) site.

To gain insights into the structure and function of the wheat (Triticum aestivum L.) genomes, we identified 278 ESTs related to abiotic stress (cold, heat, drought, salinity, and aluminum) from 7671 ESTs previously mapped to wheat chromosomes. Of the 278 abiotic stress related ESTs, 259 (811 loci) were assigned to chromosome deletion bins and analyzed for their distribution pattern among the 7 homoeologous chromosome groups. Distribution of abiotic stress related EST loci were not uniform throughout the different regions of the chromosomes of the 3 wheat genomes. Both the short and long arms of group 4 chromosomes showed a higher number of loci in their distal regions compared with proximal regions. Of the 811 loci, the number of mapped loci on the A, B, and D genomes were 258, 281, and 272, respectively. The highest number of abiotic stress related loci were found in homoeologous chromosome group 2 (142 loci) and the lowest number were found in group 6 (94 loci). When considering the genome-specific ESTs, the B genome showed the highest number of unique ESTs (7 loci), while none were found in the D genome. Similarly, considering homoeologous group-specific ESTs, group 2 showed the highest number with 16 unique ESTs (58 loci), followed by group 4 with 9 unique ESTs (33 loci). Many of the classified proteins fell into the biological process categories associated with metabolism, cell growth, and cell maintenance. Most of the mapped ESTs fell into the category of enzyme activity (28%), followed by binding activity (27%). Enzymes related to abiotic stress such as beta-galactosidase, peroxidase, glutathione reductase, and trehalose-6-phosphate synthase were identified. The comparison of stress-responsive ESTs with genomic sequences of rice (Oryza sativa L.) chromosomes revealed the complexities of colinearity. This bin map provides insight into the structural and functional details of wheat genomic regions in relation to abiotic stress.

Identification and expression profiles of multiple genes in Nile tilapia in response to bacterial infections

The constitutive splicing factor ASF/SF2 has been shown to affect the choice between alternative splice sites by favoring the proximal as opposed to the distal choice. HnRNP A1 antagonizes ASF/SF2 by promoting the distal choice for competing 5' splice sites. We have tested the in vivo effects of these proteins on alternative 3' splice site choices. Cotransfection of a dihydrofolate reductase-calcitonin chimeric construct togetherwith a plasmid specifying the SR protein ASF/SF2 into cells of several mammalian lines increased use of a proximal 3' splice site, resulting in the inclusion of a terminal calcitonin exon. This stimulation of 3' proximal splicing was antagonized by cotransfection with an hnRNP A1 plasmid. This effect of hnRNP A1 in promoting distal splicing was also seen in an hnRNP A1-deficient MEL cell line. A similar effect of hnRNP A1 was demonstrated with mutant hamster adenine phosphoribosyltransferase (aprt) transcripts that are normally constitutively spliced, suggesting that hnRNP A1 may be a general inhibitor of proximal splicing. Intron size also influenced splice site choice in mutant aprt transcripts, with larger introns favoring proximal splicing. These results support the idea that the ratios of particular but general splicing factors and hnRNPs play a role in alternative splicing.

Computational efforts to understand the process of pre-mRNA splicing have led to the development of various tools that can separately predict the locations of branchpoints and splice sites. Since experimental studies have shown that the majority of branchpoints are distributed just upstream of the 3' splice site, it should be possible to use this information to jointly predict the locations of the branchpoint and splice acceptor site. Here, we propose a deep neural network based sequence-to-sequence learning solution that can label each nucleotide of an input sequence as being a branchpoint or a splice site or neither. We demonstrate how mutation maps can be generated to interpret our model and use them to show that our network learns the interdependence between the branchpoint and the splice site. An analysis of these maps also reveals that our model is able to learn about other influential sequence features such as the polypyrimidine tract and alternative splice sites and make predictions of mutations in these regions.

Identification and characterization of the murine FK506 binding protein (FKBP) 12.6 gene.