Patterns and architecture of genomic islands in marine bacteria
- Beatriz Fernández-Gómez†1,
- Antonio Fernàndez-Guerra†2,
- Emilio O Casamayor2,
- José M González3,
- Carlos Pedrós-Alió1 and
- Silvia G Acinas1Email author
© Fernández-Gómez et al.; licensee BioMed Central Ltd. 2012
Received: 27 February 2012
Accepted: 10 July 2012
Published: 29 July 2012
Genomic Islands (GIs) have key roles since they modulate the structure and size of bacterial genomes displaying a diverse set of laterally transferred genes. Despite their importance, GIs in marine bacterial genomes have not been explored systematically to uncover possible trends and to analyze their putative ecological significance.
We carried out a comprehensive analysis of GIs in 70 selected marine bacterial genomes detected with IslandViewer to explore the distribution, patterns and functional gene content in these genomic regions. We detected 438 GIs containing a total of 8152 genes. GI number per genome was strongly and positively correlated with the total GI size. In 50% of the genomes analyzed the GIs accounted for approximately 3% of the genome length, with a maximum of 12%. Interestingly, we found transposases particularly enriched within Alphaproteobacteria GIs, and site-specific recombinases in Gammaproteobacteria GIs. We described specific Homologous Recombination GIs (HR-GIs) in several genera of marine Bacteroidetes and in Shewanella strains among others. In these HR-GIs, we recurrently found conserved genes such as the β-subunit of DNA-directed RNA polymerase, regulatory sigma factors, the elongation factor Tu and ribosomal protein genes typically associated with the core genome.
Our results indicate that horizontal gene transfer mediated by phages, plasmids and other mobile genetic elements, and HR by site-specific recombinases play important roles in the mobility of clusters of genes between taxa and within closely related genomes, modulating the flexible pool of the genome. Our findings suggest that GIs may increase bacterial fitness under environmental changing conditions by acquiring novel foreign genes and/or modifying gene transcription and/or transduction.
KeywordsGenomic islands Horizontal gene transfer Homologous recombination Bacterial core genes Flexible genome Structure of genomic islands Patterns within genomic islands Marine bacteria
Bacterial comparative genomics is providing a unique opportunity to retrieve valuable information regarding genome structure, functional diversity and evolution of marine microorganisms. Bacterial genomes are dynamic entities with a conserved pool of genes at the core genome shared at different taxonomic levels, and the flexible (or adaptive) genome, with a number of taxa-specific genes that are not comparable among closely related strains[1, 2]. Horizontal gene transfer (HGT) is one of the evolutionary mechanisms enlarging the flexible genome pool of bacterial populations, facilitating their adaptation to new ecological niches[3, 4]. GIs are clusters of genes laterally transferred and associated with the flexible genome pool of prokaryotic genomes. These highly variable genome regions have been analyzed for a few bacterial taxa by comparative genome analysis. In well-known marine bacteria such as Prochlorococcus, Synechococcus, and Shewanella comparative genome analysis has revealed a substantial number of species-specific genes[5–7]. Those studies revealed an unsaturated pangenome size, reflecting the existence of new lineages and the heterogeneity among the flexible genome pool.
Species-specific genes are commonly found in GIs[1, 8]. These are important genomic regions causing significant genetic differences between closely related taxa, and they may reveal particular ecologically relevant features of the genomes[9, 10] and virus-bacteria interaction. GIs may harbor a large set of genes with different origins. However, using a hypothesis-free approach for the identification of GIs some common features can be recognized suggesting that GIs could be perceived as a superfamily of mobile elements. Genes found within GIs are very diverse: from key genes for survival in specific environments to virulence, and/or antibiotic resistance genes. In fact, GIs enriched in virulence genes and in Clustered Regularly Interspaced Palindromic Repeats (CRISPR) confer resistance to exogenous genetic elements such as plasmids and phages. Thus, GI content may hold clues about the lifestyle or survival strategies of bacteria. Another common characteristic of the GIs is the enrichment in novel genes without any orthologous groups detected in the Clusters of Orthologous Groups (COG) database or any other known functional gene families.
Extensive literature exists on GIs in pathogenic bacterial strains (referred to as pathogenic islands) where their relevance is known for antibiotic resistance or virulence stages[1, 16, 17]. In environmental microorganisms GIs have been associated with the presence of catabolic pathways for organic pollutants, thus conferring adaptive traits in some Pseudomonas strains. Another ecological feature associated with GIs is the presence of genes for magnetite biomineralization in what is called the magnetosome island in the alphaproteobacterium Magnetospirillum gryphiswaldense, or secondary metabolism in marine Actinobacteria strains. Another case is the acquisition of a capsular polysaccharide biosynthesis gene cluster by the non-pathogenic soil inhabitant Burkholderia thailandensis with similar characteristics to the virulence gene cluster of the pathogenic Burkholderia pseudomallei (responsible for the melioidosis disease). In Cyanobacteria, GIs from several strains of Prochlorococcus marinus[5, 10] and Synechococcus strains have been reported. Also, the GIs of two freshwater filamentous toxin-producing cyanobacteria were found with diverse comparative approaches. For Gammaproteobacteria, GIs were described for the high pressure adapted Photobacterium profundum SS9 strain, the marine coastal Vibrio vulnificus, Alteromonas macleodii and Shewanella baltica strains. In Alphaproteobacteria, GIs were found in SAR11 (Candidatus Pelagibacter ubique strain HTCC1062) referred to as hypervariable regions in the original study. In aquatic Bacteroidetes GIs were described in Salinibacter ruber, a very abundant bacterium in solar salterns. Finally, virulence genes of typical pathogenic island were reported in marine bacteria genomes in a comparative study.
However, GIs in marine bacterial genomes have not been explored systematically and a comparative analysis is still lacking. Bacteria can adapt to different light regimes or to attach to organic matter particles for example, allowing niche differentiation and coexistence of different species. Therefore, analyses of GIs of marine bacterial genomes may reveal genes for adaptation to different ecological niches. In this study, we carried out a comprehensive analysis of GIs in 70 selected marine bacterial genomes that represented abundant and ecologically relevant bacteria in the ocean. We assembled a database of 8152 genes found in GIs of marine bacteria and screened it for possible patterns and clues about the ecological relevance of GIs in marine bacteria.
Results and discussion
Accuracy of GI prediction: previous (control) vs. GIs detected in this study
Comparison of the GIs of eight marine bacteria referred to as Control Genomes where GIs were available in previous studies and the GIs predicted by this study for the same genomes by IslandViewer
Previous studies GIs
Comparison analyses (Previous vs. IslandViewer GIs)
Number of GIs
Total GIs length detected in control genomes (kb)
#GIs present (+)
#GIs absent (−)
Total GIs length with IslandViewer prediction (kb)*
P. marinus str. MIT9312 1
S. sp. RCC307 2
S. sp. WH7803 2
S. sp. CC9605 2
S. sp. CC9311 2
A. macleodii “deep ecotype” 3
S. ruber DSM13855 4
S. ruber M8 4
The GIs of previous studies are shown in blue in Figure1 together with the GIs detected by IslandViewer in red. The percentage of GIs detected ranged from 27% in Synechococcus RCC307 to 100% in both Salinibacter strains. Another discrepancy was that some areas were detected as GIs that were not annotated as such in the genomes. In most cases there were only one or two “extra GIs” per genome, although there were five in Alteromonas and four in Synechococcus CC9605. In all cases, however, these extra GIs were very short (Figure1), usually smaller than 8 kb, and therefore these GIs were not included in our final dataset. In terms of the length of DNA in GIs (in kb) IslandViewer detected approximately 20–60% of the length in manually annotated GIs with an average of 36% (Table1). These percentages were smaller than those found in a previous test based on 118 bacterial genomes (mostly pathogenic bacteria with many closely related genomes available) also using IslandViewer (88% on average)[32, 33]. However, both precision (average 73%) and sensitivity (average 64%) were very good when compared with the different methods explored in (see their Table 1).
Additionally, we compared the functional gene annotation in the GIs that were detected by both systems (previous studies and this study; Table 1). We did two types of comparisons. In the first one, we considered each genome separately and, in the second one, we considered all the control genomes pooled together. Obviously, for this purpose, only genes with clearly assigned gene categories (GO) could be considered. Thus, out of all the detected genes, the HPs genes were discarded for this comparison (Additional file1). The number of genes ranged from 51 to 225 in the manually annotated subset and from 16 to 116 in the automatically annotated subset. Fischer’s exact tests revealed no significant differences in the proportion of genes in each GO category between the two data subsets. Finally, we pooled all the annotated genes in previous published GIs of the eight control genomes (1065 genes), and all the predicted genes in the GIs for the same genomes (397 genes) detected by IslandViewer. We only found three specific GO terms with significantly different distributions in both datasets. One GO term related to photosynthesis (GO:0015979) was found underrepresented in the IslandViewer annotated database. This was probably due to the smaller percentage of GIs detected in some of the cyanobacterial strains, such as Synechococcus sp. CC9311 and RCC307 or Prochlorococcus marinus MIT9312. Two other GO terms were overrepresented in the IslandViewer annotated data set. These were related to DNA recombination (GO:0006310) and DNA binding (GO:003677). This was likely due to the fact that IslandViewer detects many GIs based on the existence of mobile genes (Additional file2). Likely, this was also the cause of the “extra GIs”, since isolated mobile elements tend to be ignored during manual annotation unless they are an important objective for the researcher (and usually they are not). Thus, IslandViewer should be more efficient at detecting these GIs than previous studies based on a single approach.
In conclusion, taking into account that our GIs prediction is not exhaustive and will be missing some of the true GIs, the important conclusion for the present work is that the functional analyses of genes in GIs from a large number of marine bacterial genomes will uncover valid patterns and ecologically relevant information in this flexible genome pool.
Quantitative importance of GIs in marine bacterial genomes
The 70 selected marine bacterial genomes represent the four major prokaryotic taxa in the ocean: Cyanobacteria (16 genomes), Gammaproteobacteria (17), Alphaproteobacteria (16) and Bacteroidetes (21) (Additional file3: Table S3). These four bacterial taxa account for up to 80% of the total marine bacterioplankton. Bacteroidetes genomes included 14 Flavobacteria and 7 non-marine Bacteroidetes (Bacteroides spp.) used as out-groups. Several genomes of closely related bacterial strains from each phylogenetic group were included to investigate the rate of variability of GIs at intra-specific level and explore their relevance as main contributors to strain-specific genes. IslandViewer detects GIs ≥8 kb although only those ≥9.5 kb were used to compile our database for further analyses to be consistent with previous published studies[6, 9, 25, 37].
Interestingly, high intra-specific variability in GIs size was observed in some members of each group. Two Synechococcus strains, for example, with genomes of 2.2 and 2.6 Mb respectively, showed very different GI ratios (15 and 175 kb respectively). Similarly, small differences in genome size between two Shewanella baltica strains (MR-4 and OS155, with 4.7 and 5.12 Mb respectively) contrasted with marked differences in their GIs ratios (3.7 and 6.3% respectively) (Additional file3 and Additional file4).
Architecture of marine bacterial GIs
As a common characteristic we found that 70% of the detected GIs contained MGE, mostly transposases, conjugative transposons, integrons or phage integrase-related genes in accordance with previous GIs studies[39, 40]. In addition, we observed that at least 27% of the GIs were flanked by or contained tRNAs, probably acting as the integration sites for GIs[41, 42].
Some of the genes present in these particular HR-GIs encode ribosomal proteins, which are known to be highly expressed, usually with a sequence composition different from the rest of the genome. As a consequence, these genome fragments might appear as false-positive predictions of GIs if sequence composition bias (% GC content) were used as the only criterion to identify GIs. IslandViewer integrated the three most accurate GI prediction programs[15, 32, 45], each using different approaches to predict GIs and, in effect, both HR-GIs were detected by more than one tool (see rows in yellow in Additional file2). However, we looked for additional evidence that these gene-cassettes were not false positives, that is, that they were in a true GI.
For this purpose, phylogenetic trees were reconstructed with sequences from 20 Bacteroidetes genomes based on RpoB and EF-Tu gene sequences, which are found in HR1-GIs, as well as the 16S rRNA. If these GIs were false-positives, we would expect the phylogenies of RpoB and EF-Tu genes to match that of 16S rRNA. If these were true GIs subject to HR, however, we would expect somewhat different phylogenies for 16S rRNA, and RpoB and EF-Tu genes (Additional file6). Although the general topology among Flavobacteria, Sphingobacteria and Bacteroides branches was conserved with the three genes, several discrepancies could be detected between the 16S rRNA phylogeny and those of the other two genes ( see black triangles and circles in Additional file6). This is in accordance with the two functional genes following similar evolutionary trends and belonging to a GI.
These HR1-GIs related to transcription and its regulation and translation processes if they have been adequately integrated might be beneficial under particular conditions. We observed this HR1-GI next to a cluster of ribosomal protein genes and rpoA as occurs for many Shewanella strains or next to the TonB-dependent receptors as is the case for some marine Bacteroidetes (data not shown). This strategic location may favor an increased level of synthesis of proteins required at critical moments or at transitions to different lifestyles, as suggested for marine Bacteroidetes.
Functional annotation of the prokaryotic GIs
The possibility to move clusters of genes associated with transcription and their regulation (presence of β subunit of DNA-directed RNA polymerases, elongation factor Tu, sigma factors, transcription termination factors) and translation between closely related genomes and/or different genera may be beneficial for bacterial fitness under changing environmental conditions when an increased level of the transcription and synthesis of certain proteins may be needed.
It seems that lateral transfer genes related to protein-protein interactions may take up several million of years to be established into the regulatory network of the host. Thus, the gene clusters detected in GIs may represent ancient transfer events. If they have been integrated next to other related regulatory proteins or transcriptional factors, their selection might have been favored. Additionally, IS/transposase genes located nearby and within GIs may have a key role to activate transcription of those genes by either introducing complete or partial promoters located within the element itself, by disrupting another gene that may inhibit transcription or by inserting foreign genes into positions where they become regulated by endogenous promoters.
Differences in GI gene content among marine bacterial classes
To find out whether there were differences in the functional categories found within GIs of the four different bacterial classes, we compared the representation of GO terms in each main phylogenetic group with the remaining dataset by paired Fisher´s Exact Tests (Additional file8). There were significant differences in all cases, indicating that each bacterial class had a different set of functions preferentially represented in their GIs. Our results have to be interpreted with caution since differences in GIs number between organisms might partly be due to variations in efficiency of GI prediction between organisms, due for instance to other factors causing a bias in sequence composition such as a difference in gene expression level.
Gene Ontology (GO) terms enrichment analyses of GIs in 4 main phylogenetic groups
Single test p-value
Internal side of plasma membrane
Transition metal ion binding
Cobalamin biosynthetic process
Electron transport chain
Glucose metabolic process
Proton-transporting two-sector ATPase complex
Plasma membrane ATP synthesis coupled proton transport
Hydrogen ion transporting ATP synthase activity
Histidine biosynthetic process
Intracellular membrane-bounded organelle
Hydrolase activity, acting on acid anhydrides
Site-specific recombinase activity
RNA-directed DNA polymerase activity
Acyl-CoA dehydrogenase activity
ATPase activity, coupled to transmembrane
movement of ions
DNA-directed RNA polymerase activity
Transcription elongation regulator activity
Positive regulation of RNA elongation from
RNA pol. II promoter
Translation factor activity, nucleic acid binding
Structural constituent of ribosome
Small ribosomal subunit
Large ribosomal subunit
DNA topoisomerase type I activity
Fisher Exact Test was used with distinct statistically methods: False Discovery Rate control (FDR), Family Wise Error Rate (FWER) and the p-value without multiple testing corrections (single test p-value). Only the most specific GO terms overrepresented in each bacterial group using FDR with statistical significance (**p-value <0.01; ***p-value <0.001) are shown. Five bacterial groups were analyzed: Cyanobacteria, Gammaproteobacteria, Alphaproteobacteria; Flavobacteria; and non-marine Bacteroidetes.
Alphaproteobacteria GIs were enriched (six GO terms) in genes related to transposases, DNA transposition activity, and motility. Surprisingly, RNA-directed DNA-polymerase activity (a signature for the presence of retrovirus prophages) and ribosome assemblage, related to a high number of ribosomal proteins, were also specifically enriched in marine Alphaproteobacteria genomes. The two GO terms enriched in Gammaproteobacteria included genes associated to site-specific recombinases and ligase activity associated with DNA mobility and rearrangements.
Flavobacteria were specifically enriched in eight GO terms with genes associated to ATPase and GTPase activity and, interestingly, in processes related to DNA-directed RNA polymerases activity, transcription, and its regulation processes (Table2). GIs of non-marine Bacteroidetes were enriched in seven GO terms, involved in translation processes and the structure of the ribosome (many ribosomal proteins associated with the large and small ribosomal subunits) and rRNA and tRNA binding.
Therefore, each bacterial class contains a different set of genes in their GIs, suggesting a different ecological strategy played by their GIs. This will be analyzed in the next section.
Biologically relevant genes within marine bacterial GIs
Transposases and integrases (category 12 in Figure10) can modify the structure of the genome through the transfer of DNA sequences to new locations within or between genomes. Recently, it has been reported that transposases are the most abundant and ubiquitous genes in nature. We found a total of 675 genes related to transposases such as IS elements or transposons representing about 8.2% of the total database. These genes were overrepresented in Alphaproteobacteria within GIs (see significance test in Table 2) with almost a 30% of total transposases (194 genes) within this group (data not shown). Also, ribosomal proteins (category 3 in Figure10) accounted for 3% of all genes in our dataset (253 out of 8152 genes) and were very abundant in Alphaproteobacteria and Gammaproteobacteria genomes (also in non-marine Bacteroidetes) with almost a 25% and 22% respectively (data not shown). Specifically, marine genomes with more than 10 gene copies were Sulfitobacter sp. EE-36 (32 copies) and Ruegeria pomeroyi DSS-3 (18 copies) in the Alphaproteobacteria, and Shewanella baltica OS155 (26 copies) and Vibrio cholerae O395 (29 copies) in the Gammaproteobacteria.
Virulence gene clusters (category 14 in Figure10) were found in all main taxa except for Alphaproteobacteria with the highest number for Vibrio cholerae O395 with 19 copies corresponding to the well known TCP (Toxin-Coregulated Pilus) located in one of the pathogenic island organized as a prophage. Finally, the presence of secretion system proteins (category 16 in Figure10) was found in all taxa but marine Flavobacteria. Specifically, Thalassiobium sp. R2A62 exhibited seven copies of Type I secretion system proteins and at least 3 copies of type III secretion system were found in Shewanella baltica OS155. These virulence associated secretion system proteins have been found in other marine genomes. Particularly, type IV and type VI secretions system genes were recurrent for Alpha- and Gammaproteobacteria genomes respectively.
Our GI dataset also included genes involved in protection from bacteriophages such as DNA modification restriction systems (type I, II and III, category 6 in Figure10). These systems are sequence-specific restriction enzymes, also called restriction endonucleases that have been known for a long time to act as a protection from foreign DNA, such as bacteriophages. We detected 99 genes linked to restriction modification systems spread along all taxa but specially represented in: Anabaena variabilis ATCC 29413 with five copies, Roseobacter denitrificans OCh 114 (10 copies), Ruegeria pomeroyi DSS-3 (eight copies) and Psychrobacter cryohalolentis K5 (11 of them). Finally, three marine Bacteroidetes had five copies each (Cytophaga hutchinsonii ATCC 33406, Kordia algicida OT-1 and Robiginitalea biformata HTCC2501) (Figure10). Interestingly, most of the restriction-modification system genes in our GI dataset were of the type I (the most complex) with at least 36 genes, but we also found some representatives of type II and III (data not shown).
Another mechanism proposed to confer resistance from phage and possibly from other mobile elements is the presence of CRISPR systems with their associated cas genes[63–65]. These genetics elements have been identified in approximately 40% and 90% of Bacteria and Archaea genomes, respectively. Recent research has shown that CRISPR systems could be primarily transferred by horizontal gene transfer and can be found overrepresented within GIs. Although we did not find many of them (category 13 in Figure10) associated to our marine prokaryotic GI dataset, we found cas genes in Anabaena variabilis (four copies), in Rhodobacter sphaeroides KD131 (seven copies) and four genes in Leeuwenhoekiella blandensis MED217. Some of these CRISPR systems were already in the CRISPRdb (http://crispr.u-psud.fr/crispr/) in the chromosome of these genomes although they were not associated with GIs and no CRISPR had been previously identified in Leeuwenhoekiella blandensis MED217.
Other categories were restricted to a few genomes and/or phylogenetic groups. This was the case of photosynthetic genes (category 1 in Figure10) within cyanobacterial taxa. Photosystem I subunits and photosynthesis antenna proteins as well as electron transporter systems (ATP synthases) or ferredoxins were detected in Cyanobacteria GIs. Photosynthesis genes within GIs have been well described in Prochlorococcus and Synechococcus strains[5, 6, 9] but we also found them in Synechocystis sp. PCC 6803 with three phycobilisome linker proteins in the 14.6 kb GI, as well as in Anabaena variabilis ATCC 29413 with two photosystem I subunit proteins and one ferredoxin, and in Nostoc sp. PCC7120 with eight ATP synthase subunits, two phycobilisome linker proteins and another two allophycocyanin alpha/beta subunits. These photosynthetic genes are linked to relevant physiological characteristics of these photoautotrophic bacteria. The acquisition of these genes by GIs may provide specific light niche adaptations to specific strains, underlying the need for analyzing their GIs to fully understand the ecology of Cyanobacteria.
In addition, we detected cell motility (flagellum) genes (category 10 in Figure10) constrained basically to Roseobacter denitrificans OCh 114 and Roseobacter sp. MED193 with 20 and 11 copies, respectively, and Shewanella baltica OS155 with eight. Cell motility by flagella is an important physiological trait that allows bacteria to move towards favorable environmental conditions, form biofilms and/or acquire nutrients. Genes to reconstruct the flagellum structure can include more than 50 but only 24 are considered to be the core set and are present in most flagellated bacterial taxa. Some of theses genes can be acquired through HGT events as described in Photobacterium profundum SS9. In this organism a cluster of genes involved in the lateral flagellar synthesis was present in the GI and absent in a closely related strain, suggesting that it could have been horizontally transferred. Accordingly, 43 genes within two Roseobacter genomes and two Gammaproteobacteria might have followed the same fate. Genes related to the flagellar basal rod, body, ring, and hook or flagellin proteins were repeatedly found in these GIs. In Roseobacter denitrificans OCh 114, 14 of these flagellar genes were concentrated in two GIs of 14.1 kb and 16 kb with 12 and 8 genes respectively. These GIs were flanked by transposase, integrase or phage- integrase genes, while in Roseobacter sp. MED193 these genes were in a single GI 12 kb long, also flanked by phage-integrase genes. In addition, Shewanella baltica OS155 had eight flagellar related genes in the GI of 11.5 kb and interestingly, a chemotaxis protein gene was located within the same GI. Moreover, one of the GIs (18.7 kb) of Alteromonas macleodii "deep ecotype" displayed eight flagellar protein gene in GI8 (20 kb) previously reported in reference. We investigated whether these flagellar protein genes present in our GIs dataset were also present in the genome but we found different genes for the flagellum structure in the chromosome (data not shown). Finally, conjugative transposon genes (category 12 in Figure10) were found only in a few taxa. Two Alphaproteobacteria, Jannaschia sp. CCS1 and Sphingomonas wittichii RW1 displayed six and seven genes related to these conjugative transposons, and the Bacteroidetes Flavobacterium johnsoniae UW101 had four copies.
It is highly probable that many genes within GIs are positively selected due to the potential benefits of these genes for the life-style of their host. It has been shown that GIs of Prochlorococcus strains displayed differential expression under light and nutrient stress conditions. Also, Prochlorococcus GIs contain genes related to the attachment of virions to the host cell surface and those GIs have an important role in the viruses-host coexistence. Moreover, one of the GIs with heavy metal resistance genes detected in Alteromonas macleodii "deep ecotype" conferred to this strain higher resistance to mercury and zinc concentrations. In a different context, in the Actinobacteria Salinispora arenicola, orthologs within GIs showed evidence of positive selection compared with the non-island genes (7.6% vs. 1.6%). Also, the hyperhalophilic bacterium Salinibacter ruber strains M8 and M31 displayed 40 strains-specific genes present in their GIs with a ratio of substitution rates at non-synonymous and synonymous sites >1 (dN/dS >1) in which 25 of them were HPs. Consequently, there is some evidence for the adaptive significance of GI genes among environmental bacteria, but the extent and effect on diversification mechanisms in marine bacterial taxa is still unclear.
GIs were present in most of the marine bacterial genomes analyzed. Our results indicated that both horizontal gene transfer by phages, plasmids and MGE and HR play an important role for the mobility of clusters of genes between taxa and within closely related genomes, thus modulating the flexible pool of the genome. Our findings provide insights into the possible role of GIs to increase bacterial fitness under changing environmental conditions by providing, not only novel foreign genes, but also modulating their transcription, regulation, and/or transduction. The potential role that GIs have in rearranging the structure, and increasing the diversity, of marine bacterial genomes is emphasized by the results presented here. We observed that some GIs were intimately associated with the physiology and ecology of the microorganisms but we also found some relevant conserved genes in theory linked to the core genome. These results would reinforce the need to establish a pangenome concept for marine bacterial species wherein GIs would be crucial to fully understand the ecology and evolution of marine bacteria in the ocean. Exploring the mechanisms maintaining and selecting GIs is the next logical step to gain insights into the evolutionary processes shaping marine bacterial genomes.
GI prediction and database construction
Seventy prokaryotic genomes were analyzed in this study. GIs of 53 genomes were obtained at the time of our analyses (February 2010) directly from IslandViewer database (http://www.pathogenomics.sfu.ca/islandviewer). IslandViewer is a web-based interface that integrates several methods for identification and visualization of GIs: IslandPick, IslandPath-DIMOB and SIGI-HMM. IslandPick is a comparative GIs prediction method that requires phylogenetically related genomes to be available for the comparison. SIGI-HMM measures codon usage and IslandPath the abnormal sequence composition or the presence of genes related to mobile elements to identify possible GIs. For a recent review of Bioinformatics approaches to detect GIs see. The remaining 17 genomes (eight Alphaproteobacteria and nine marine Bacteroidetes) were downloaded from the National Center for Biotechnology Information (NCBI;http://www.ncbi.nlm.nih.gov) and the J. Craig Venter Institute (http://www.jcvi.org/). Fourteen of these 17 genomes were not closed. The genomes that were not available in IslandViewer were uploaded and the GIs predicted by IslandPick by selecting closely related genomes (at least three genomes when possible) plus one reference distant genome. IslandPick GIs prediction detects also the GIs that overlap with the other two GIs predictors, IslandPath-DIMOB and SIGI-HMM and this information was also integrated in our database. The 438 GIs detected are shown in Additional file2 including predictor methods used to detect them and some features such as the presence of MGE like plasmids, transposases, integrons, conjugative transposons or phages. Most of the GIs were detected by at least two of the three methods integrated in IslandViewer. In addition, manual refining of these GIs was carried out in Artemis and Integrated Microbial Genomes (IMG) genome browser, by including the presence of tRNA within or flanking the GIs detected. GIs of the 70 bacterial genomes were exported in csv format and all proteins in a fasta file. The genome selection criteria used were: (i) presence of marine genomes with pre-calculated GIs within IslandViewer, (ii) availability when possible of at least three closely related genomes, and (ii) the widest possible taxonomic representation of ecologically relevant marine bacteria.
Maximum likelihood (ML) phylogenetic analyses of selected genomes were carried out on full-length 16 S rRNA gene sequences, and the elongation factor Tu (EF-Tu) gene and the ß subunit of the RNA polymerase (RpoB) on amino acid sequences. The sequences were aligned using MAFFT [v.6.857] with the algorithm E-INS-I. An additional more stringent alignment was constructed by removing ambiguously aligned sites using Gblocks as well as visual examination. Phylogenies were constructed using ML as implemented in RAxML [v.7.2.8] with GTR nucleotide substitution model for the 16S rRNA sequences and the BLOSUM62 amino acid substitution matrix for the EF-Tu and RpoB sequences. The trees generated were visualized and edited in Interactive Tree Of Life.
GIs analyses in previous studies (control) vs. this study
We selected eight genomes found in the public sequence databases as control genomes, in which the GIs had been described, to test the accuracy of our approach for GI prediction in marine genomes even when 3 closely related genomes were not always available for comparison. We calculated several parameters such as % overlap, precision, recall (sensitivity) as detailed in Langille et al. shown in Table 1. The GIs of control genomes were verified by different approaches (see Table 1). Comparison of the control and automatically predicted GIs was carried out using the genomic display software CIRCOS.
Functional annotation of GIs
Gene sequences found in the predicted GIs were extracted and stored in a flat file-type database. The potential protein domains were analyzed using the package HMMER 3.0 against the PFAM 24.0 database. Clusters of orthologous groups of proteins (COGs) for characterization of the proteins were carried out using the rpsblast program bundled in the NCBI BLAST package with an E-value of 1E-5 as threshold. In addition, Blast2GO (B2G) was used to run the functional annotation of the extracted sequences with an E-value of 1E-20 and cut-off identity in their amino acid sequences of 55%[77, 78]. Gene ontologies and EC number from KEGG pathways were retrieved to identify the main biological processes, molecular functions, and cellular components present in the GIs.
Statistical and GIs comparison analyses
Both total genome peptides and peptides within GIs were classified by BLASTP hits to the NCBI's Cluster of Orthologous Genes (COG) as described in with cutoffs of E-value ≤1E-5, identity ≥ 30 and coverage ≥ 50%. HPs were considered as those with no hits to the COG database. To estimate protein functions that were overrepresented in different taxonomic groups we used the Gossip package implemented in the Bioinformatics annotation tool Blast2GO. This package uses the Fisher´s Exact Test with multiple testing corrections and, since multiple categories are examined simultaneously, the Benjamini and Hochberg False Discovery Rate correction (FDR) for multiple testing was determined for all functional category analyses. We considered p-values smaller than 0.01 to be significant (p <0.01). Fisher´s Exact Test (FT) was used to explore putative significant differences at three comparative levels: (i) the functional gene annotation of the combined GIs of each control genome and their corresponding detected automatically GIs, (ii) the combined GIs dataset for the eight control genomes and their corresponding automatically detected GIs and (iii) the comparative gene enrichment analyses among the bacterial taxa of Cyanobacteria, Gammaproteobacteria, Alphaproteobacteria and marine and non-marine Bacteroidetes to detect functional categories enriched within each taxa.
In addition, we assigned genes within GIs to 16 biological categories defined by us based on the most representative GO terms with ecologically relevance and related to: photosynthesis (1) (photosystem, antenna proteins and electron transport system), energy metabolism enzymes (2), ribosomal proteins (3), hydrolysis (4), polysaccharide biosynthesis (5), DNA restriction modification system (type I, II & III) (6), DNA-directed RNA polymerases (7), transporters (ABC and multridrug/metal resistence) (8), two component system (9), cell motility (flagellum and chemotaxis) (10), stress response (heat shock or chaperone proteins) (11), MGE (conjugative transposon, integrases, phage integrases, transposon Tn21/Tn7) (12), CRISPRs (13), virulence (14), TA toxins (plasmid killer system) (15), and secretion systems proteins (type I, II and III) (16). A matrix was built showing the number of genes within GI assigned to each of the 16 biological categories (Additional file9). GO terms enrichment analyses and the matrix of ecological gene categories, were implemented in the iTOL software for visualization (http://itol.embl.de/).
SGA conceived and designed the analyses. BF-G, AF-G, JMG and SGA performed the analyses and analyzed the data. All authors helped in interpreting the data. SGA wrote the paper with significant contributions from all authors with special mention to CPA. All authors read and approved the final manuscript.
Horizontal Gene Transfer
Clustered Regularly Interspaced Palindromic Repeat
Cluster of Orthologous Groups
Mobile Genetic Element
β-subunit of DNA-directed RNA polymerase
elongation factor Tu
National Center for Biotechnology Information
Integrated Microbial Genomes
We thank Guillem Salazar for his help on statistical analyses of the data and Ramiro Logares for critical reading. We thank two anonymous reviewers for their helpful comments. BF-G was a recipient of an I3P grant from CSIC and AF-G was supported by grant CONSOLIDER-INGENIO2010 GRACCIE CSD2007-00067 from the Spanish Ministry of Science and Innovation to EOC. SGA was supported by a RyC contract from Spanish Ministry of Science and Innovation and CONES 2010–0036 from the Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR). This research was supported mostly by grant MICRODIVERSITY (CGL2008-00762/BOS) to SGA. Other projects involved were CGL2009-13318-CO2-01/BOS to EOC and MarineGems (CTM2010-20361) from the Spanish Ministry of Science and Innovation to JMG and CPA.
- Hacker J, Carniel E: Ecological fitness, genomic islands and bacterial pathogenicity. EMBO Rep. 2001, 2 (5): 376-381.PubMed CentralView ArticlePubMedGoogle Scholar
- Ochman H, Lerat E, Daubin V: Examining bacterial species under the specter of gene transfer and exchange. Proc Natl Acad Sci USA. 2005, 102 (1): 6595-6599. 10.1073/pnas.0502035102.PubMed CentralView ArticlePubMedGoogle Scholar
- Doolittle WF: Lateral genomics. Trends Cell Biol. 1999, 9 (12): M5-M8. 10.1016/S0962-8924(99)01664-5.View ArticlePubMedGoogle Scholar
- Boucher Y, Douady CJ, Papke RT, Walsh DA, Boudreau MER, Nesbø CL, Case RJ, Doolittle WF: Lateral gene transfer and the origins of prokaryotic groups. Annu Rev Genet. 2003, 37 (1): 283-328. 10.1146/annurev.genet.37.050503.084247.View ArticlePubMedGoogle Scholar
- Kettler GC, Martiny AC, Huang K, Zucker J, Coleman ML, Rodrigue S, Chen F, Lapidus A, Ferriera S, Johnson J, Steglich C, Church GM, Richardson P, Chisholm SW: Patterns and implications of gene gain and loss in the evolution of Prochlorococcus. PLoS Genet. 2007, 3 (12): e231-10.1371/journal.pgen.0030231.PubMed CentralView ArticlePubMedGoogle Scholar
- Dufresne A, Ostrowski M, Scanlan D, Garczarek L, Mazard S, Palenik B, Paulsen I, de Marsac N, Wincker P, Dossat C, Ferriera S, Johnson J, Post AF, Hess WR, Partenshky R: Unraveling the genomic mosaic of a ubiquitous genus of marine cyanobacteria. Genome Biol. 2008, 9 (5): R90-10.1186/gb-2008-9-5-r90.PubMed CentralView ArticlePubMedGoogle Scholar
- Konstantinidis KT, Serres MH, Romine MF, Rodrigues JLM, Auchtung J, McCue L-A, Lipton MS, Obraztsova A, Giometti CS, Nealson KH, Fredrickson JK, Tiedje JM: Comparative systems biology across an evolutionary gradient within the Shewanella genus. Proc Natl Acad Sci USA. 2009, 106 (37): 15909-15914. 10.1073/pnas.0902000106.PubMed CentralView ArticlePubMedGoogle Scholar
- Hacker J, Kaper JB: Pathogenicity islands and the evolution of microbes. Annu Rev Microbiol. 2000, 54 (1): 641-679. 10.1146/annurev.micro.54.1.641.View ArticlePubMedGoogle Scholar
- Coleman M, Sullivan M, Martiny A, Steglich C, Barry K, DeLong E, Chisholm S: Genomic islands and the ecology and evolution of Prochlorococcus. Science. 2006, 311 (5768): 1768-1770. 10.1126/science.1122050.View ArticlePubMedGoogle Scholar
- Cuadros-Orellana S, Martín-Cuadrado A-B, Legault B, D'Auria G, Zhaxybayeva O, Papke RT, Rodríguez-Valera F: Genomic plasticity in prokaryotes: the case of the square haloarchaeon. ISMEJ. 2007, 1 (3): 235-245. 10.1038/ismej.2007.35.View ArticleGoogle Scholar
- Avrani S, Wurtzel O, Sharon I, Sorek R, Lindell D: Genomic island variability facilitates Prochlorococcus-virus coexistence. Nature. 2011, 474 (7353): 604-608. 10.1038/nature10172.View ArticlePubMedGoogle Scholar
- Vernikos GS, Parkhill J: Resolving the structural features of genomic islands: a machine learning approach. Genome Res. 2008, 18 (2): 331-342. 10.1101/gr.7004508.PubMed CentralView ArticlePubMedGoogle Scholar
- Ho Sui SJ, Fedynak A, Hsiao WWL, Langille MGI, Brinkman FSL: The association of virulence factors with genomic islands. PLoS One. 2009, 4 (12): e8094-10.1371/journal.pone.0008094.PubMed CentralView ArticlePubMedGoogle Scholar
- Read TD, Ussery DW: Opening the pan-genomics box. Curr Opin Microbiol. 2006, 9: 496-198. 10.1016/j.mib.2006.08.010.View ArticleGoogle Scholar
- Hsiao W, Ung K, Aeschliman D, Bryan J, Finlay B, Brinkman F: Evidence of a large novel gene pool associated with prokaryotic genomic islands. PLoS Genet. 2005, 1 (5): e62-10.1371/journal.pgen.0010062.PubMed CentralView ArticlePubMedGoogle Scholar
- Schmidt H, Hensel M: Pathogenicity islands in bacterial pathogenesis. Clin Microbiol Rev. 2004, 17 (1): 14-56. 10.1128/CMR.17.1.14-56.2004.PubMed CentralView ArticlePubMedGoogle Scholar
- Gal-Mor O, Finlay BB: Pathogenicity islands: a molecular toolbox for bacterial virulence. Cell Microbiol. 2006, 8 (11): 1707-1719. 10.1111/j.1462-5822.2006.00794.x.View ArticlePubMedGoogle Scholar
- van der Meer JR, Sentchilo V: Genomic islands and the evolution of catabolic pathways in bacteria. Curr Opin Biotech. 2003, 14 (3): 248-254. 10.1016/S0958-1669(03)00058-2.View ArticlePubMedGoogle Scholar
- Ullrich S, Kube M, Schubbe S, Reinhardt R, Schuler D: A hypervariable 130-kilobase genomic region of Magnetospirillum gryphiswaldense comprises a magnetosome island which undergoes frequent rearrangements during stationary growth. J Bacteriol. 2005, 187 (21): 7176-7184. 10.1128/JB.187.21.7176-7184.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- Penn K, Jenkins C, Nett M, Udwary DW, Gontang EA, McGlinchey RP, Foster B, Lapidus A, Podell S, Allen EE, Moore BS, Jensen PR: Genomic islands link secondary metabolism to functional adaptation in marine Actinobacteria. ISMEJ. 2009, 3 (10): 1193-1203. 10.1038/ismej.2009.58.View ArticleGoogle Scholar
- Sim BMQ, Chantratita N, Ooi WF, Nandi T, Tewhey R, Wuthiekanun V, Thaipadungpanit J, Tumapa S, Ariyaratne P, Sung WK, Sem XH, Chua HH, Ramnarayanan K, Lin CH, Liu Y, Feil EJ, Glass MB, Tan G, Peacock SJ, Tan P: Genomic acquisition of a capsular polysaccharide virulence cluster by non-pathogenic Burkholderia isolates. Genome Biol. 2010, 11 (8): R89-10.1186/gb-2010-11-8-r89.PubMed CentralView ArticlePubMedGoogle Scholar
- Stucken K, John U, Cembella A, Murillo AA, Soto-Liebe K, Fuentes-Valdés JJ, Friedel M, Plominsky AM, Vásquez M, Glöckner G: The smallest known genomes of multicellular and toxic Cyanobacteria: Comparison, minimal gene sets for linked traits and the evolutionary implications. PLoS One. 2010, 5 (2): e9235-10.1371/journal.pone.0009235.PubMed CentralView ArticlePubMedGoogle Scholar
- Campanaro S, Vezzi A, Vitulo N, Lauro F, D'Angelo M, Simonato F, Cestaro A, Malacrida G, Bertoloni G, Valle G, Bartlett DH: Laterally transferred elements and high pressure adaptation in Photobacterium profundum strains. BMC Genomics. 2005, 6 (1): 122-10.1186/1471-2164-6-122.PubMed CentralView ArticlePubMedGoogle Scholar
- Cohen ALV, Oliver JD, DePaola A, Feil EJ, Boyd FE: Emergence of a virulent clade of Vibrio vulnificus and correlation with the presence of a 33-kilobase genomic island. Appl Environ Microbiol. 2007, 73 (17): 5553-5565. 10.1128/AEM.00635-07.PubMed CentralView ArticlePubMedGoogle Scholar
- Ivars-Martínez E, Martin-Cuadrado A-B, D'Auria G, Mira A, Ferriera S, Johnson J, Friedman R, Rodríguez-Valera F: Comparative genomics of two ecotypes of the marine planktonic copiotroph Alteromonas macleodii suggests alternative lifestyles associated with different kinds of particulate organic matter. ISMEJ. 2008, 2 (12): 1194-1212. 10.1038/ismej.2008.74.View ArticleGoogle Scholar
- Caro-Quintero A, Deng J, Auchtung J, Brettar I, Hofle MG, Klappenbach J, Konstantinidis KT: Unprecedented levels of horizontal gene transfer among spatially co-occurring Shewanella bacteria from the Baltic Sea. ISMEJ. 2011, 5 (1): 131-140. 10.1038/ismej.2010.93.View ArticleGoogle Scholar
- Wilhelm L, Tripp HJ, Givan S, Smith D, Giovannoni S: Natural variation in SAR11 marine bacterioplankton genomes inferred from metagenomic data. Biol Direct. 2007, 2 (1): 27-10.1186/1745-6150-2-27.PubMed CentralView ArticlePubMedGoogle Scholar
- Pasic L, Rodríguez-Mueller B, Martin-Cuadrado A-B, Mira A, Rohwer F, Rodríguez-Valera F: Metagenomic islands of hyperhalophiles: the case of Salinibacter ruber. BMC Genomics. 2009, 10 (1): 570-10.1186/1471-2164-10-570.PubMed CentralView ArticlePubMedGoogle Scholar
- Persson OP, Pinhassi J, Riemann L, Marklund B-I, Rhen M, Normark S, González JM, Hagström Å: High abundance of virulence gene homologues in marine bacteria. Environ Microbiol. 2009, 11 (6): 1348-1357. 10.1111/j.1462-2920.2008.01861.x.PubMed CentralView ArticlePubMedGoogle Scholar
- Stomp M, Huisman J, de Jongh F, Veraart AJ, Gerla D, Rijkeboer M, Ibelings BW, Wollenzien UIA, Stal LJ: Adaptive divergence in pigment composition promotes phytoplankton biodiversity. Nature. 2004, 432 (7013): 104-107. 10.1038/nature03044.View ArticlePubMedGoogle Scholar
- Gómez-Consarnau L, Fernàndez-Guerra A, Goesmann A, Pedrós-Alió C: Genomics of the proteorhodopsin-containing marine flavobacterium Dokdonia sp. Strain MED134. Appl Environ Microbiol. 2011, 77 (24): 8676-8686. 10.1128/AEM.06152-11.PubMed CentralView ArticlePubMedGoogle Scholar
- Langille MGI, Hsiao WWL, Brinkman FSL: Detecting genomic islands using bioinformatics approaches. Nat Rev Micro. 2010, 8 (5): 373-382. 10.1038/nrmicro2350.View ArticleGoogle Scholar
- Langille M, Hsiao W, Brinkman F: Evaluation of genomic island predictors using a comparative genomics approach. BMC Bioinformatics. 2008, 9 (1): 329-10.1186/1471-2105-9-329.PubMed CentralView ArticlePubMedGoogle Scholar
- Langille MGI, Brinkman FSL: IslandViewer: an integrated interface for computational identification and visualization of genomic islands. Bioinformatics. 2009, 25: 664-665. 10.1093/bioinformatics/btp030.PubMed CentralView ArticlePubMedGoogle Scholar
- Hsiao W, Wan I, Jones S, Brinkman F: IslandPath: aiding detection of genomic islands in prokaryotes. Bioinformatics. 2003, 19 (3): 418-420. 10.1093/bioinformatics/btg004.View ArticlePubMedGoogle Scholar
- Barberán A, Casamayor EO: Global phylogenetic community structure and ß-diversity patterns in surface bacterioplankton metacommunities. Aquat Microb Ecol. 2010, 59 (1): 1-10.View ArticleGoogle Scholar
- Peña A, Teeling H, Huerta-Cepas J, Santos F, Yarza P, Brito-Echeverria J, Lucio M, Schmitt-Kopplin P, Meseguer I, Schenowitz C, Dossat C, Barbe V, Dopazo J, Rosselló-Mora R, Schüler M, Glöckner FO, Amann R, Gabaldón T, Antón J: Fine-scale evolution: genomic, phenotypic and ecological differentiation in two coexisting Salinibacter ruber strains. ISMEJ. 2010, 4 (7): 882-895. 10.1038/ismej.2010.6.View ArticleGoogle Scholar
- Ochman H, Lawrence JG, Groisman EA: Lateral gene transfer and the nature of bacterial innovation. Nature. 2000, 405 (6784): 299-304. 10.1038/35012500.View ArticlePubMedGoogle Scholar
- Dobrindt U, Hochhut B, Hentschel U, Hacker J: Genomic islands in pathogenic and environmental microorganisms. Nat Rev Micro. 2004, 2 (5): 414-424. 10.1038/nrmicro884.View ArticleGoogle Scholar
- Juhas M, van der Meer JR, Gaillard M, Harding R, Hood D, Crook D: Genomic islands: tools of bacterial horizontal gene transfer and evolution. FEMS Microbiol Rev. 2009, 33 (2): 376-393. 10.1111/j.1574-6976.2008.00136.x.PubMed CentralView ArticlePubMedGoogle Scholar
- Reiter W-D, Palm P, Yeats S: Transfer RNA genes frequently serve as integration sites for prokaryotic genetic elements. Nucleic Acids Res. 1989, 17 (5): 1907-1914. 10.1093/nar/17.5.1907.PubMed CentralView ArticlePubMedGoogle Scholar
- Williams KP: Integration sites for genetic elements in prokaryotic tRNA and tmRNA genes: sublocation preference of integrase subfamilies. Nucleic Acids Res. 2002, 30 (4): 866-875. 10.1093/nar/30.4.866.PubMed CentralView ArticlePubMedGoogle Scholar
- Qin Q-L, Zhang X-Y, Wang X-M, Liu G-M, Chen X-L, Xie B-B, Dang H-Y, Zhou B-C, Yu J, Zhang Y-Z: The complete genome of Zunongwangia profunda SM-A87 reveals its adaptation to the deep-sea environment and ecological role in sedimentary organic nitrogen degradation. BMC Genomics. 2010, 11 (1): 247-10.1186/1471-2164-11-247.PubMed CentralView ArticlePubMedGoogle Scholar
- Karlin S: Detecting anomalous gene clusters and pathogenicity islands in diverse bacterial genomes. Trends Microbiol. 2001, 9 (7): 335-343. 10.1016/S0966-842X(01)02079-0.View ArticlePubMedGoogle Scholar
- Waack S, Keller O, Asper R, Brodag T, Damm C, Fricke W, Surovcik K, Meinicke P, Merkl R: Score-based prediction of genomic islands in prokaryotic genomes using hidden Markov models. BMC Bioinformatics. 2006, 7 (1): 142-10.1186/1471-2105-7-142.PubMed CentralView ArticlePubMedGoogle Scholar
- Casey J, Daly C, Fitzgerald GF: Chromosomal integration of plasmid DNA by homologous recombination in Enterococcus faecalis and Lactococcus lactis subsp. lactis hosts harboring Tn919. Appl Environ Microbiol. 1991, 57 (9): 2677-2682.PubMed CentralPubMedGoogle Scholar
- Dikow R: Genome-level homology and phylogeny of Shewanella (Gammaproteobacteria: lteromonadales: Shewanellaceae). BMC Genomics. 2011, 12 (1): 237-10.1186/1471-2164-12-237.PubMed CentralView ArticlePubMedGoogle Scholar
- Venkateswaran K, Moser DP, Dollhopf ME, Lies DP, Saffarini DA, MacGregor BJ, Ringelberg DB, White DC, Nishijima M, Sano H, Burghardt J, Stackebrandt E, Nealson KH: Polyphasic taxonomy of the genus Shewanella and description of Shewanella oneidensis sp. nov. Int J Syst Bacteriol. 1999, 49 (2): 705-724. 10.1099/00207713-49-2-705.View ArticlePubMedGoogle Scholar
- Fraser C, Alm EJ, Polz MF, Spratt BG, Hanage WP: The bacterial species challenge: making sense of genetic and ecological diversity. Science. 2009, 323 (5915): 741-746. 10.1126/science.1159388.View ArticlePubMedGoogle Scholar
- Thomas CM, Nielsen KM: Mechanisms of, and barriers to, horizontal gene transfer between bacteria. Nat Rev Micro. 2005, 3 (9): 711-721. 10.1038/nrmicro1234.View ArticleGoogle Scholar
- Hanage WP, Spratt BG, Turner KME, Fraser C: Modelling bacterial speciation. Philos Trans R Soc Lond B Biol Sci. 2006, 361 (1475): 2039-2044. 10.1098/rstb.2006.1926.PubMed CentralView ArticlePubMedGoogle Scholar
- Fraser C, Hanage WP, Spratt BG: Recombination and the nature of bacterial speciation. Science. 2007, 315 (5811): 476-480. 10.1126/science.1127573.PubMed CentralView ArticlePubMedGoogle Scholar
- Papke RT, Koenig JE, Rodríguez-Valera F, Doolittle WF: Frequent recombination in a saltern population of Halorubrum. Science. 2004, 306 (5703): 1928-1929.PubMedGoogle Scholar
- Zehr JP, Bench SR, Mondragon EA, McCarren J, DeLong EF: Low genomic diversity in tropical oceanic N2-fixing cyanobacteria. Proc Natl Acad Sci USA. 2007, 104 (45): 17807-17812. 10.1073/pnas.0701017104.PubMed CentralView ArticlePubMedGoogle Scholar
- Lercher MJ, Pál C: Integration of horizontally transferred genes into regulatory interaction networks takes many million years. Mol Biol Evol. 2008, 25 (3): 559-567. 10.1093/molbev/msm283.View ArticlePubMedGoogle Scholar
- Syvanen M: The evolutionary implications of mobile genetic elements. Annu Rev Genet. 1984, 18: 271-293. 10.1146/annurev.ge.18.120184.001415.View ArticlePubMedGoogle Scholar
- Kasak L, Horak R, Nurk A, Talvik K, Kivisaar M: Regulation of the catechol 1,2-dioxygenase- and phenol monooxygenase-encoding pheBA operon in Pseudomonas putida PaW85. J Bacteriol. 1993, 175 (24): 8038-8042.PubMed CentralPubMedGoogle Scholar
- Karlin S, Mrázek J, Campbell AM: Codon usages in different gene classes of the Escherichia coli genome. Mol Microbiol. 1998, 29 (6): 1341-1355. 10.1046/j.1365-2958.1998.01008.x.View ArticlePubMedGoogle Scholar
- Rice PA, Baker TA: Comparative architecture of transposase and integrase complexes. Nat Struct Mol Biol. 2001, 8 (4): 302-307. 10.1038/86166.View ArticleGoogle Scholar
- Aziz RK, Breitbart M, Edwards RA: Transposases are the most abundant, most ubiquitous genes in nature. Nucleic Acids Res. 2010, 38 (13): 4207-4217. 10.1093/nar/gkq140.PubMed CentralView ArticlePubMedGoogle Scholar
- Karaolis DKR, Somara S, Maneval DR, Johnson JA, Kaper JB: A bacteriophage encoding a pathogenicity island, a type-IV pilus and a phage receptor in cholera bacteria. Nature. 1999, 399 (6734): 375-379. 10.1038/20715.View ArticlePubMedGoogle Scholar
- Arber W, Linn S: DNA modification and restriction. Annu Rev Biochem. 1969, 38: 467-500. 10.1146/annurev.bi.38.070169.002343.View ArticlePubMedGoogle Scholar
- Mojica FJM, Díez-Villaseñor C, Soria E, Juez G: Biological significance of a family of regularly spaced repeats in the genomes of Archaea, Bacteria and mitochondria. Mol Microbiol. 2000, 36 (1): 244-246. 10.1046/j.1365-2958.2000.01838.x.View ArticlePubMedGoogle Scholar
- Barrangou R, Fremaux C, Deveau H, Richards M, Boyaval P, Moineau S, Romero DA, Horvath P: CRISPR provides acquired resistance against viruses in prokaryotes. Science. 2007, 315 (5819): 1709-1712. 10.1126/science.1138140.View ArticlePubMedGoogle Scholar
- Sorek R, Kunin V, Hugenholtz P: CRISPR – a widespread system that provides acquired resistance against phages in Bacteria and Archaea. Nat Rev Micro. 2008, 6 (3): 181-186. 10.1038/nrmicro1793.View ArticleGoogle Scholar
- Grissa I, Vergnaud G, Pourcel C: The CRISPRdb database and tools to display CRISPRs and to generate dictionaries of spacers and repeats. BMC Bioinformatics. 2007, 8 (1): 172-10.1186/1471-2105-8-172.PubMed CentralView ArticlePubMedGoogle Scholar
- Liu R, Ochman H: Stepwise formation of the bacterial flagellar system. Proc Natl Acad Sci USA. 2007, 104 (17): 7116-7121. 10.1073/pnas.0700266104.PubMed CentralView ArticlePubMedGoogle Scholar
- Katoh K, Toh H: Recent developments in the MAFFT multiple sequence alignment program. Brief Bioinform. 2008, 9 (4): 286-298. 10.1093/bib/bbn013.View ArticlePubMedGoogle Scholar
- Castresana J: Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Bio Evol. 2000, 17 (4): 540-552. 10.1093/oxfordjournals.molbev.a026334.View ArticleGoogle Scholar
- Stamatakis A, Hoover P, Rougemont J: A rapid bootstrap algorithm for the RAxML web servers. Syst Biol. 2008, 57 (5): 758-771. 10.1080/10635150802429642.View ArticlePubMedGoogle Scholar
- Letunic I, Bork P: Interactive Tree Of Life (iTOL): an online tool for phylogenetic tree display and annotation. Bioinformatics. 2007, 23 (1): 127-128. 10.1093/bioinformatics/btl529.View ArticlePubMedGoogle Scholar
- Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, Jones SJ, Marra MA: Circos: an information aesthetic for comparative genomics. Genome Res. 2009, 19 (9): 1639-1645. 10.1101/gr.092759.109.PubMed CentralView ArticlePubMedGoogle Scholar
- Eddy SR: Profile hidden Markov models. Bioinformatics. 1998, 14 (9): 755-763. 10.1093/bioinformatics/14.9.755.View ArticlePubMedGoogle Scholar
- Finn RD, Mistry J, Tate J, Coggill P, Heger A, Pollington JE, Gavin OL, Gunasekaran P, Ceric G, Forslund K, Eddy SR, Sonnhammer ELL, Bateman A: The Pfam protein families database. Nucleic Acids Res. 2010, 38 (suppl 1): D211-D222.PubMed CentralView ArticlePubMedGoogle Scholar
- Tatusov RL, Koonin EV, Lipman DJ: A genomic perspective on protein families. Science. 1997, 278 (5338): 631-637. 10.1126/science.278.5338.631.View ArticlePubMedGoogle Scholar
- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol. 1990, 215 (3): 403-410.View ArticlePubMedGoogle Scholar
- Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, Robles M: Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics. 2005, 21 (18): 3674-3676. 10.1093/bioinformatics/bti610.View ArticlePubMedGoogle Scholar
- Gotz S, Garcia-Gomez JM, Terol J, Williams TD, Nagaraj SH, Nueda MJ, Robles M, Talon M, Dopazo J, Conesa A: High-throughput functional annotation and data mining with the Blast2GO suite. Nucleic Acids Res. 2008, 36 (10): 3420-3435. 10.1093/nar/gkn176.PubMed CentralView ArticlePubMedGoogle Scholar
- Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G: Gene ontology: tool for the unification of biology. The gene ontology consortium. Nature Genet. 2000, 25 (1): 25-29. 10.1038/75556.PubMed CentralView ArticlePubMedGoogle Scholar
- Kanehisa M, Goto S: KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000, 28 (1): 27-30. 10.1093/nar/28.1.27.PubMed CentralView ArticlePubMedGoogle Scholar
- Lauro FM, McDougald D, Thomas T, Williams TJ, Egan S, Rice S, DeMaere MZ, Ting L, Ertan H, Johnson J, Ferriera S, Lapidus A, Anderson I, Kyrpides N, Munk AC, Detter C, Han CS, Brown MV, Robb FT, Kjelleberg S, Cavicchioli R: The genomic basis of trophic strategy in marine bacteria. Proc Natl Acad Sci USA. 2009, 106 (37): 15527-15533. 10.1073/pnas.0903507106.PubMed CentralView ArticlePubMedGoogle Scholar
- Blüthgen N, Brand K, Cajavec B, Swat M, Herzel H, Beule D: Biological profiling of gene groups utilizing gene ontology. Genome Inform. 2005, 16 (1): 106-115.PubMedGoogle Scholar