The sequence asymmetry of the Escherichia coli chromosome appears to be independent of strand or function and may be evolutionarily conserved.

I have examined potential determinants of the asymmetric distribution of nucleotide sequences in the genome of Escherichia coli as cataloged in GenBank release 44. I have used the frequency of occurrence of all possible tetranucleotides in a given sequence catalog or derivative as a comparative measure of asymmetry. The GenBank-cataloged strand and its complement show statistically similar (not complementary) distributions. The distribution is statistically similar in comparisons between the protein coding subset and the total genome, the coding subset and selected non-coding genes, the coding subset and the remainder of the DNA, and the coding subset and stable RNA sequences. I have compared the distribution in the genome of E. coli with the distributions found in the cataloged genomes of Salmonella typhimurium, Bacillus subtilis, and of coliphages lambda and T7. The distribution summed in both strands of the cataloged DNA differs statistically only in comparisons with lytic bacteriophage T7 because only the two strands of T7 show statistically dissimilar distributions. Despite similarities in tetranucleotide distribution, the pattern of codon complementarity in B. subtilis is different than that documented for E. coli. Thus, sequence asymmetry does not seem related to specific DNA function or to documented similarities or differences in codon bias. The sequence asymmetry of the E. coli genome may thus reflect a hitherto unsuspected pattern impressed on both strands of DNA which is or can be packaged into bacterial genomes.     

An automated annotation tool for genomic DNA sequences using GeneScan and BLAST

Genomic sequence data are often available well before the annotated sequence is published. We present a method for analysis of genomic DNA to identify coding sequences using the GeneScan algorithm and characterize these resultant sequences by BLAST. The routines are used to develop a system for automated annotation of genome DNA sequences.     

A fractal method to distinguish coding and non-coding sequences in a complete genome based on a number sequence representation

A fractal method to distinguish coding and non-coding sequences in a complete genome is proposed, based on different statistical behaviors between these two kinds of sequences. We first propose a number sequence representation of DNA sequences. Multifractal analysis is then performed on the measure representation of the obtained number sequence. The three exponents C(-1), C1 and C2 are selected from the result of multifractal analysis. Each DNA may be represented by a point in the three-dimensional space generated by these three-component vectors. It is shown that points corresponding to coding and non-coding sequences in the complete genome of many prokaryotes are roughly distributed in different regions. Fisher's discriminant algorithm can be used to separate these two regions in the spanned space. If the point (C(-1),C1,C2) for a DNA sequence is situated in the region corresponding to coding sequences, the sequence is discriminated as a coding sequence; otherwise, the sequence is classified as a non-coding one. For all 51 prokaryotes we considered , the average discriminant accuracies pc,pnc,qc and qnc reach 72.28%, 84.65%, 72.53% and 84.18%, respectively.     

Gene content and density in banana (<Emphasis Type="Italic">Musa acuminata</Emphasis>) as revealed by genomic sequencing of BAC clones

The complete sequence of Musa acuminata bacterial artificial chromosome (BAC) clones is presented and, consequently, the first analysis of the banana genome organization. One clone (MuH9) is 82,723 bp long with an overall G+C content of 38.2%. Twelve putative protein-coding sequences were identified, representing a gene density of one per 6.9 kb, which is slightly less than that previously reported for Arabidopsis but similar to rice. One coding sequence was identified as a partial M. acuminata malate synthase, while the remaining sequences showed a similarity to predicted or hypothetical proteins identified in genome sequence data. A second BAC clone (MuG9) is 73,268 bp long with an overall G+C content of 38.5%. Only seven putative coding regions were discovered, representing a gene density of only one gene per 10.5 kb, which is strikingly lower than that of the first BAC. One coding sequence showed significant homology to the soybean ribonucleotide reductase (large subunit). A transition point between coding regions and repeated sequences was found at approximately 45 kb, separating the coding upstream BAC end from its downstream end that mainly contained transposon-like sequences and regions similar to known repetitive sequences of M. acuminata. This gene organization resembles Gramineae genome sequences, where genes are clustered in gene-rich regions separated by gene-poor DNA containing abundant transposons.Communicated by J.S. Heslop-Harrison     

Assembly-free quantification of vagrant DNA inserts

Inserts of DNA from extranuclear sources, such as organelles and microbes, are common in eukaryote nuclear genomes. However, sequence similarity between the nuclear and extranuclear DNA, and a history of multiple insertions, make the assembly of these regions challenging. Consequently, the number, sequence and location of these vagrant DNAs cannot be reliably inferred from the genome assemblies of most organisms. We introduce two statistical methods to estimate the abundance of nuclear inserts even in the absence of a nuclear genome assembly. The first (intercept method) only requires low-coverage (<1×) sequencing data, as commonly generated for population studies of organellar and ribosomal DNAs. The second method additionally requires that a subset of the individuals carry extranuclear DNA with diverged genotypes. We validated our intercept method using simulations and by re-estimating the frequency of human NUMTs (nuclear mitochondrial inserts). We then applied it to the grasshopper Podisma pedestris, exceptional for both its large genome size and reports of numerous NUMT inserts, estimating that NUMTs make up 0.056% of the nuclear genome, equivalent to >500 times the mitochondrial genome size. We also re-analysed a museomics data set of the parrot Psephotellus varius, obtaining an estimate of only 0.0043%, in line with reports from other species of bird. Our study demonstrates the utility of low-coverage high-throughput sequencing data for the quantification of nuclear vagrant DNAs. Beyond quantifying organellar inserts, these methods could also be used on endosymbiont-derived sequences. We provide an R implementation of our methods called “vagrantDNA” and code to simulate test data sets.     

Organizational heterogeneity of vertebrate genomes

Genomes of higher eukaryotes are mosaics of segments with various structural, functional, and evolutionary properties. The availability of whole-genome sequences allows the investigation of their structure as "texts" using different statistical and computational methods. One such method, referred to as Compositional Spectra (CS) analysis, is based on scoring the occurrences of fixed-length oligonucleotides (k-mers) in the target DNA sequence. CS analysis allows generating species- or region-specific characteristics of the genome, regardless of their length and the presence of coding DNA. In this study, we consider the heterogeneity of vertebrate genomes as a joint effect of regional variation in sequence organization superimposed on the differences in nucleotide composition. We estimated compositional and organizational heterogeneity of genome and chromosome sequences separately and found that both heterogeneity types vary widely among genomes as well as among chromosomes in all investigated taxonomic groups. The high correspondence of heterogeneity scores obtained on three genome fractions, coding, repetitive, and the remaining part of the noncoding DNA (the genome dark matter--GDM) allows the assumption that CS-heterogeneity may have functional relevance to genome regulation. Of special interest for such interpretation is the fact that natural GDM sequences display the highest deviation from the corresponding reshuffled sequences.     

Detection of protein coding sequences using a mixture model for local protein amino acid sequence.

Locating protein coding regions in genomic DNA is a critical step in accessing the information generated by large scale sequencing projects. Current methods for gene detection depend on statistical measures of content differences between coding and noncoding DNA in addition to the recognition of promoters, splice sites, and other regulatory sites. Here we explore the potential value of recurrent amino acid sequence patterns 3-19 amino acids in length as a content statistic for use in gene finding approaches. A finite mixture model incorporating these patterns can partially discriminate protein sequences which have no (detectable) known homologs from randomized versions of these sequences, and from short (< or = 50 amino acids) non-coding segments extracted from the S. cerevisiea genome. The mixture model derived scores for a collection of human exons were not correlated with the GENSCAN scores, suggesting that the addition of our protein pattern recognition module to current gene recognition programs may improve their performance.     

Genomic DNA sequence comparison between two inbred soybean cyst nematode biotypes facilitated by massively parallel 454 micro-bead sequencing

Heterodera glycines, the soybean cyst nematode (SCN), is a damaging agricultural pest that could be effectively managed if critical phenotypes, such as virulence and host range could be understood. While SCN is amenable to genetic analysis, lack of DNA sequence data prevents the use of such methods to study this pathogen. Fortunately, new methods of DNA sequencing that produced large amounts of data and permit whole genome comparative analyses have become available. In this study, 400 million bases of genomic DNA sequence were collected from two inbred biotypes of SCN using 454 micro-bead DNA sequencing. Comparisons to a BAC, sequenced by Sanger sequencing, showed that the micro-bead sequences could identify low and high copy number regions within the BAC. Potential single nucleotide polymorphisms (SNPs) between the two SCN biotypes were identified by comparing the two sets of sequences. Selected resequencing revealed that up to 84% of the SNPs were correct. We conclude that the quality of the micro-bead sequence data was sufficient for de novo SNP identification and should be applicable to organisms with similar genome sizes and complexities. The SNPs identified will be an important starting point in associating phenotypes with specific regions of the SCN genome.     

The contribution of slippage-like processes to genome evolution

Simple sequences present in long (>30 kb) sequences representative of the single-copy genome of five species (Homo sapiens, Caenorhabditis elegans Saccharomyces cerevisiae, E. coli, and Mycobacterium leprae) have been analyzed. A close relationship was observed between genome size and the overall level of sequence repetition. This suggested that the incorporation of simple sequences had accompanied increases of genome size during evolution. Densities of simple sequence motifs were higher in noncoding regions than in coding regions in eukaryotes but not in eubacteria. All five genomes showed very biased frequency distributions of simple sequence motifs in all species, particularly in eukaryotes where AAA and TTT predominated. Interspecific comparisons showed that noncoding sequences in eukaryotes showed highly significantly similar frequency distributions of simple sequence motifs but this was not true of coding sequences. ANOVA of the frequency distributions of simple sequence motifs indicated strong contributions from motif base composition and repeat unit length, but much of the variation remained unexplained by these parameters. The sequence composition of simple sequences therefore appears to reflect both underlying sequence biases in slippage-like processes and the action of selection. Frequency distributions of simple sequence motifs in coding sequences correlated weakly or not at all with those in noncoding sequences. Selection on coding sequences to eliminate undesirable sequences may therefore have been strong, particularly in the human lineage.     

Prevalence of quadruplexes in the human genome

Guanine-rich DNA sequences of a particular form have the ability to fold into four-stranded structures called G-quadruplexes. In this paper, we present a working rule to predict which primary sequences can form this structure, and describe a search algorithm to identify such sequences in genomic DNA. We count the number of quadruplexes found in the human genome and compare that with the figure predicted by modelling DNA as a Bernoulli stream or as a Markov chain, using windows of various sizes. We demonstrate that the distribution of loop lengths is significantly different from what would be expected in a random case, providing an indication of the number of potentially relevant quadruplex-forming sequences. In particular, we show that there is a significant repression of quadruplexes in the coding strand of exonic regions, which suggests that quadruplex-forming patterns are disfavoured in sequences that will form RNA.     

An investigation of genomic base distribution

While veritable oceans of ink have been spilled over the base distributions within genes, the literature is virtually silent on large scale intra genomic base distribution. To address this issue, we have examined approximately 3400 chromosomal sequences from approximately 2000 entire genomes-including DNA and RNA, single- and double-stranded, coding and non-coding genomes. For each sequence the mean, variance, skewness, and kurtosis for each base were computed along with the genome base composition. The main findings are: (1) there is no simple relationship between these statistics and the base composition of the genome, (2) in non-viral genomes, base distribution is non-uniform, (3) base distribution in non-eukaryotic genomes obeys a number of simple rules, (4) these rules are not dependent on the presence of coding sequences, (5) bacterial genomes in particular are unusually compliant with these rules, and (6) eukaryotes have a unique pattern of base distribution.     

Similarity Estimation Between DNA Sequences Based on Local Pattern Histograms of Binary Images

Graphical representation of DNA sequences is one of the most popular techniques for alignment-free sequence comparison. Here, we propose a new method for the feature extraction of DNA sequences represented by binary images, by estimating the similarity between DNA sequences using the frequency histograms of local bitmap patterns of images. Our method shows linear time complexity for the length of DNA sequences, which is practical even when long sequences, such as whole genome sequences, are compared. We tested five distance measures for the estimation of sequence similarities, and found that the histogram intersection and Manhattan distance are the most appropriate ones for phylogenetic analyses.     

Applications of hidden Markov models for characterization of homologous DNA sequences with a common gene.

Identifying and characterizing the structure in genome sequences is one of the principal challenges in modern molecular biology, and comparative genomics offers a powerful tool. In this paper, we introduce a hidden Markov model that allows a comparative analysis of multiple sequences related by a phylogenetic tree, and we present an efficient method for estimating the parameters of the model. The model integrates structure prediction methods for one sequence, statistical multiple alignment methods, and phylogenetic information. This unified model is particularly useful for a detailed characterization of DNA sequences with a common gene. We illustrate the model on a variety of homologous sequences.     

Gene identification in novel eukaryotic genomes by self-training algorithm

Finding 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 in DNA 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.     

ALU-ring elements in the primate genomes

Elucidation of complete nucleotide sequence of the human has revealed that coding sequences that store the information needed to synthesize functional proteins, occupy only 2% of the genomic region. The remaining 98%, barring few regulatory sequences, has been referred to as non-functional or junk DNA and consists of many kinds of repeat elements. In fact, human genome is the most repeat rich genome sequenced so far, in which more than half of the region is occupied by such sequences. Determination of significance of these repeats in the human genome has become the focus of many studies all over the world, especially after genome sequencing did not reveal any significant difference in coding regions between lower eukaryotes and human. In this article, we have focused on Alu repeats that are primate specific elements with many interesting biological properties. Moreover, these are the repeats with highest copy number in the human genome. We have highlighted different facets of their interaction with the genome and changing paradigms regarding their role in genome organization.     

Identification of coding and non-coding sequences using local Holder exponent formalism

MOTIVATION: Accurate prediction of genes in genomes has always been a challenging task for bioinformaticians and computational biologists. The discovery of existence of distinct scaling relations in coding and non-coding sequences has led to new perspectives in the understanding of the DNA sequences. This has motivated us to exploit the differences in the local singularity distributions for characterization and classification of coding and non-coding sequences. RESULTS: The local singularity density distribution in the coding and non-coding sequences of four genomes was first estimated using the wavelet transform modulus maxima methodology. Support vector machines classifier was then trained with the extracted features. The trained classifier is able to provide an average test accuracy of 97.7%. The local singularity features in a DNA sequence can be exploited for successful identification of coding and non-coding sequences. CONTACT: Available on request from bd.kulkarni@ncl.res.in.     

An analysis of the genome of Ciona intestinalis

An analysis by CsCl density gradient centrifugation has shown that, at a fragment size of about 100 kb, the DNA of a urochordate, Ciona intestinalis, is remarkably homogeneous in base composition. Localization of 16 coding sequences from C. intestinalis, chosen so as to cover the distribution range of all available coding sequences for this organism, showed a nearly symmetrical distribution almost coinciding with the DNA distribution. Both distributions are remarkably different from those found in vertebrates, which are skewed towards high GC levels (to a greater extent in warm-blooded vertebrates). In order to account for this change in genome organization, we propose a working hypothesis that can be tested. Basically, we suggest that the genome duplication that occurred between urochordates and fishes was accompanied by a preferential integration of transposons in one compartment of the genome, which was made gene-poor (by lowering gene density) compared to the rest. Since the gene-poor compartment (the 'empty quarter') is characterized by a lower level of gene expression compared to the gene-rich compartment (the 'genome core') in the vertebrate genome, we further suggest, as a working hypothesis, that a compartmentalization according to gene expression already existed in urochordates.     

Using proteomics to mine genome sequences

We present a method for mining unannotated or annotated genome sequences with proteomic data to identify open reading frames. The region of a genome coding for a protein sequence is identified by using information from the analysis of proteins and peptides with MALDI-TOF mass spectrometry. The raw genome sequence or any unassembled contigs of an organism are theoretically cleaved into a number of equal sized but overlapping fragments, and these are then translated in all six frames into a series of virtual proteins. Each virtual protein is then subjected to a theoretical enzymatic digestion. Standard proteomic sample preparation methods are used to separate, array, and digest the proteins of interest to peptides. The masses of the resulting peptides are measured using mass spectrometry and compared to the theoretical peptide masses of the virtual proteins. The region of the genome responsible for coding for a particular protein can then be identified when there are a large number of hits between peptides from the protein and peptides from the virtual protein. The method makes no assumptions about the location of a protein in a particular gene sequence or the positions or types of start and stop codons. To illustrate this approach, all 773 proteins of Pseudomonas aeruginosa contained in SWISS-PROT were used to theoretically test the method and optimize parameters. Increasing the size of the virtual proteins results in an overall improvement in the ability to detect the coding region, at the cost of decreasing the sensitivity of the method for smaller proteins. Increasing the minimum number of matching peptides, lowering the mass error tolerance, or increasing the signal-to-noise ratio of the simulated mass spectrum, improves the ability to detect coding regions. The method is further demonstrated on experimental data from Mycobacterium tuberculosis and is also shown to work with eukaryotic organisms (e.g., Homo sapiens).     

Repeat sequence interspersion in coding DNA of peas does not reflect that in total pea DNA

The pattern of sequence organization in the regions of the pea genome near sequences coding for mRNA differs significantly from that in total DNA. Interspersion of repeated and single copy sequences is so extensive that 85% of 1300 nucleotide-long fragments contain highly repetitive sequences (about 5000 copies per haploid genome). However, data presented here demonstrate that sequences which code for mRNA are enriched in the small fraction of fragments which do not contain these highly repetitive sequences. Thus, in contrast to the great majority of other sequences in the genome, most mRNA coding sequences are not located within 1300 nucleotides of highly repetitive elements. Moreover, our data indicate that those repeats (if any) which are closely associated with mRNA coding sequences belong to low copy number families characterized by an unusually low degree of sequence divergence.Abbreviations NT nucleotides - NTP nucleotide pairs - C_ot the product of molar concentration of DNA nucleotides and time of incubation (mol s/L) - T_m the temperature at which half of the nucleotides are unpaired - T_m,i the temperature at which half of the complementary strands are completely separated - PIPES 1, 4, Piperazinediethane sulfonic acid - PB an equimolar mixture of NaH₂PO₄ and Na₂HPO₄ (pH 6.8).