Person:
Borodovsky,
Mark
Borodovsky,
Mark
Permanent Link
Associated Organization(s)
Organizational Unit
ORCID
ArchiveSpace Name Record
Publication Search Results
Now showing
1 - 6 of 6
-
ItemImproving accuracy of gene prediction programs of the genemark family by means of genome segmentation(Georgia Institute of Technology, 2009-06-22) Borodovsky, Mark
-
ItemGene prediction in novel fungal genomes using an ab initio algorithm with unsupervised training(Georgia Institute of Technology, 2008-12) Ter-Hovhannisyan,Vardges ; Lomsadze, Alexandre ; Chernoff, Yury O. ; Borodovsky, MarkWe describe a new ab initio algorithm, GeneMark-ES version 2, that identifies protein-coding genes in fungal genomes. The algorithm does not require a predetermined training set to estimate parameters of the underlying hidden Markov model (HMM). Instead, the anonymous genomic sequence in question is used as an input for iterative unsupervised training. The algorithm extends our previously developed method tested on genomes of Arabidopsis thaliana, Caenorhabditis elegans, and Drosophila melanogaster. To better reflect features of fungal gene organization, we enhanced the intron submodel to accommodate sequences with and without branch point sites. This design enables the algorithm to work equally well for species with the kinds of variations in splicing mechanisms seen in the fungal phyla Ascomycota, Basidiomycota, and Zygomycota. Upon self-training, the intron submodel switches on in several steps to reach its full complexity. We demonstrate that the algorithm accuracy, both at the exon and the whole gene level, is favorably compared to the accuracy of gene finders that employ supervised training. Application of the new method to known fungal genomes indicates substantial improvement over existing annotations. By eliminating the effort necessary to build comprehensive training sets, the new algorithm can streamline and accelerate the process of annotation in a large number of fungal genome sequencing projects
-
ItemEvaluating the protein coding potential of exonized transposable element sequences(Georgia Institute of Technology, 2007-11-26) Piriyapongsa, Jittima ; Rutledge, Mark T. ; Patel, Sanil ; Borodovsky, Mark ; Jordan, I. KingBackground: Transposable element (TE) sequences, once thought to be merely selfish or parasitic members of the genomic community, have been shown to contribute a wide variety of functional sequences to their host genomes. Analysis of complete genome sequences have turned up numerous cases where TE sequences have been incorporated as exons into mRNAs, and it is widely assumed that such 'exonized' TEs encode protein sequences. However, the extent to which TE-derived sequences actually encode proteins is unknown and a matter of some controversy. We have tried to address this outstanding issue from two perspectives: i-by evaluating ascertainment biases related to the search methods used to uncover TE-derived protein coding sequences (CDS) and ii-through a probabilistic codon-frequency based analysis of the protein coding potential of TE-derived exons. Results: We compared the ability of three classes of sequence similarity search methods to detect TE-derived sequences among data sets of experimentally characterized proteins: 1-a profile-based hidden Markov model (HMM) approach, 2-BLAST methods and 3-RepeatMasker. Profile based methods are more sensitive and more selective than the other methods evaluated. However, the application of profile-based search methods to the detection of TE-derived sequences among well-curated experimentally characterized protein data sets did not turn up many more cases than had been previously detected and nowhere near as many cases as recent genome-wide searches have. We observed that the different search methods used were complementary in the sense that they yielded largely non-overlapping sets of hits and differed in their ability to recover known cases of TE-derived CDS. The probabilistic analysis of TE-derived exon sequences indicates that these sequences have low protein coding potential on average. In particular, non-autonomous TEs that do not encode protein sequences, such as Alu elements, are frequently exonized but unlikely to encode protein sequences. Conclusion: The exaptation of the numerous TE sequences found in exons as bona fide protein coding sequences may prove to be far less common than has been suggested by the analysis of complete genomes. We hypothesize that many exonized TE sequences actually function as post-transcriptional regulators of gene expression, rather than coding sequences, which may act through a variety of double stranded RNA related regulatory pathways. Indeed, their relatively high copy numbers and similarity to sequences dispersed throughout the genome suggests that exonized TE sequences could serve as master regulators with a wide scope of regulatory influence.
-
ItemExonization of the LTR transposable elements in human genome(Georgia Institute of Technology, 2007-08-28) Piriyapongsa, Jittima ; Polavarapu, Nalini ; Borodovsky, Mark ; McDonald, John F.Background: Retrotransposons have been shown to contribute to evolution of both structure and regulation of protein coding genes. It has been postulated that the primary mechanism by which retrotransposons contribute to structural gene evolution is through insertion into an intron or a gene flanking region, and subsequent incorporation into an exon. Results: We found that Long Terminal Repeat (LTR) retrotransposons are associated with 1,057 human genes (5.8%). In 256 cases LTR retrotransposons were observed in protein-coding regions, while 50 distinct protein coding exons in 45 genes were comprised exclusively of LTR RetroTransposon Sequence (LRTS). We go on to reconstruct the evolutionary history of an alternatively spliced exon of the Interleukin 22 receptor, alpha 2 gene (IL22RA2) derived from a sequence of retrotransposon of the Mammalian apparent LTR retrotransposons (MaLR) family. Sequencing and analysis of the homologous regions of genomes of several primates indicate that the LTR retrotransposon was inserted into the IL22RA2 gene at least prior to the divergence of Apes and Old World monkeys from a common ancestor (~25 MYA). We hypothesize that the recruitment of the part of LTR as a novel exon in great ape species occurred prior to the divergence of orangutans and humans from a common ancestor (~14 MYA) as a result of a single mutation in the proto-splice site. Conclusion: Our analysis of LRTS exonization events has shown that the patterns of LRTS distribution in human exons support the hypothesis that LRTS played a significant role in human gene evolution by providing cis-regulatory sequences; direct incorporation of LTR sequences into protein coding regions was observed less frequently. Combination of computational and experimental approaches used for tracing the history of the LTR exonization process of IL22RA2 gene presents a promising strategy that could facilitate further studies of transposon initiated gene evolution.
-
ItemImproving accuracy of gene prediction programs of the genemark family by means of genome segmentation(Georgia Institute of Technology, 2007-02-28) Borodovsky, Mark
-
ItemGene identification in novel eukaryotic genomes by self-training algorithm(Georgia Institute of Technology, 2005) Lomsadze, Alexandre ; Ter-Hovhannisyan, Vardges ; Chernoff, Yury O. ; Borodovsky, MarkFinding new protein-coding genes is one of the most important goals of eukaryotic genome sequencing projects. However, genomic organization of novel eukaryotic genomes is diverse and ab initio gene finding tools tuned up for previously studied species are rarely suitable for efficacious gene hunting inDNA sequences of a new genome. Gene identification methods based on cDNA and expressed sequence tag (EST) mapping to genomic DNA or those using alignments to closely related genomes rely either on existence of abundant cDNA and EST data and/ or availability on reference genomes. Conventional statistical ab initio methods require large training sets of validated genes for estimating gene model parameters. In practice, neither one of these types of data may be available in sufficient amount until rather late stages of the novel genome sequencing. Nevertheless, we have shown that gene finding in eukaryotic genomes could be carried out in parallel with statistical models estimation directly from yet anonymous genomic DNA. The suggested method of parallelization of gene prediction with the model parameters estimation follows the path of the iterative Viterbi training. Rounds of genomic sequence labeling into coding and non-coding regions are followed by the rounds of model parameters estimation. Several dynamically changing restrictions on the possible range of model parameters are added to filter out fluctuations in the initial steps of the algorithm that could redirect the iteration process away from the biologically relevant point in parameter space. Tests on well-studied eukaryotic genomes have shown that the new method performs comparably or better than conventional methods where the supervised model training precedes the gene prediction step. Several novel genomes have been analyzed and biologically interesting findings are discussed. Thus, a self-training algorithm that had been assumed feasible only for prokaryotic genomes has now been developed for ab initio eukaryotic gene identification.