Limits...
Dynamics of genome rearrangement in bacterial populations.

Darling AE, Miklós I, Ragan MA - PLoS Genet. (2008)

Bottom Line: The inversion phylogenies agree with results suggested by SNP patterns.We found that all inversions are shorter than expected under a neutral model, whereas inversions acting within a single replichore are much shorter than expected.We also found evidence for a canonical configuration of the origin and terminus of replication.

View Article: PubMed Central - PubMed

Affiliation: ARC Center of Excellence in Bioinformatics, The University of Queensland, St. Lucia, Queensland, Australia. darling@cs.wisc.edu

ABSTRACT
Genome structure variation has profound impacts on phenotype in organisms ranging from microbes to humans, yet little is known about how natural selection acts on genome arrangement. Pathogenic bacteria such as Yersinia pestis, which causes bubonic and pneumonic plague, often exhibit a high degree of genomic rearrangement. The recent availability of several Yersinia genomes offers an unprecedented opportunity to study the evolution of genome structure and arrangement. We introduce a set of statistical methods to study patterns of rearrangement in circular chromosomes and apply them to the Yersinia. We constructed a multiple alignment of eight Yersinia genomes using Mauve software to identify 78 conserved segments that are internally free from genome rearrangement. Based on the alignment, we applied Bayesian statistical methods to infer the phylogenetic inversion history of Yersinia. The sampling of genome arrangement reconstructions contains seven parsimonious tree topologies, each having different histories of 79 inversions. Topologies with a greater number of inversions also exist, but were sampled less frequently. The inversion phylogenies agree with results suggested by SNP patterns. We then analyzed reconstructed inversion histories to identify patterns of rearrangement. We confirm an over-representation of "symmetric inversions"-inversions with endpoints that are equally distant from the origin of chromosomal replication. Ancestral genome arrangements demonstrate moderate preference for replichore balance in Yersinia. We found that all inversions are shorter than expected under a neutral model, whereas inversions acting within a single replichore are much shorter than expected. We also found evidence for a canonical configuration of the origin and terminus of replication. Finally, breakpoint reuse analysis reveals that inversions with endpoints proximal to the origin of DNA replication are nearly three times more frequent. Our findings represent the first characterization of genome arrangement evolution in a bacterial population evolving outside laboratory conditions. Insight into the process of genomic rearrangement may further the understanding of pathogen population dynamics and selection on the architecture of circular bacterial chromosomes.

Show MeSH

Related in: MedlinePlus

Calculating expected inversion length.The expected length of within- and inter-replichore inversions can be calculated as integral averages of the function min{/x−y/,1−/x−y/} over the appropriate areas. Here, 0<b<1 is the terminus dif site. See the text for more details.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2483231&req=5

pgen-1000128-g012: Calculating expected inversion length.The expected length of within- and inter-replichore inversions can be calculated as integral averages of the function min{/x−y/,1−/x−y/} over the appropriate areas. Here, 0<b<1 is the terminus dif site. See the text for more details.

Mentions: Assume the endpoints of an inversion are in positions x and y, with x, y∈[0,1]. The inversion length can be expressed as the function min{/x−y/,1−/x−y/}, since the inversion occurs on a circular chromosme of length 1 and for any inversion longer than 0.5, a complementary inversion with shorter length exists. If we assume that the inversion endpoints are uniformly distributed, then the expected length is the integral average of the function min{/x−y/,1−/x−y/} over the appropriate area A:(9)where /A/ denotes the size of the area. In the case of within-replichore inversions, area A is the union of the two squares as delineated by the dashed line of Fig. 12, in case of inter-replichore inversions, A is the union of the two rectangles. For simplicity we suppress the full details of integration, and the resulting equations for within- and inter-replichore inversions are given in Equations 4 and 5, respectively.


Dynamics of genome rearrangement in bacterial populations.

Darling AE, Miklós I, Ragan MA - PLoS Genet. (2008)

Calculating expected inversion length.The expected length of within- and inter-replichore inversions can be calculated as integral averages of the function min{/x−y/,1−/x−y/} over the appropriate areas. Here, 0<b<1 is the terminus dif site. See the text for more details.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2483231&req=5

pgen-1000128-g012: Calculating expected inversion length.The expected length of within- and inter-replichore inversions can be calculated as integral averages of the function min{/x−y/,1−/x−y/} over the appropriate areas. Here, 0<b<1 is the terminus dif site. See the text for more details.
Mentions: Assume the endpoints of an inversion are in positions x and y, with x, y∈[0,1]. The inversion length can be expressed as the function min{/x−y/,1−/x−y/}, since the inversion occurs on a circular chromosme of length 1 and for any inversion longer than 0.5, a complementary inversion with shorter length exists. If we assume that the inversion endpoints are uniformly distributed, then the expected length is the integral average of the function min{/x−y/,1−/x−y/} over the appropriate area A:(9)where /A/ denotes the size of the area. In the case of within-replichore inversions, area A is the union of the two squares as delineated by the dashed line of Fig. 12, in case of inter-replichore inversions, A is the union of the two rectangles. For simplicity we suppress the full details of integration, and the resulting equations for within- and inter-replichore inversions are given in Equations 4 and 5, respectively.

Bottom Line: The inversion phylogenies agree with results suggested by SNP patterns.We found that all inversions are shorter than expected under a neutral model, whereas inversions acting within a single replichore are much shorter than expected.We also found evidence for a canonical configuration of the origin and terminus of replication.

View Article: PubMed Central - PubMed

Affiliation: ARC Center of Excellence in Bioinformatics, The University of Queensland, St. Lucia, Queensland, Australia. darling@cs.wisc.edu

ABSTRACT
Genome structure variation has profound impacts on phenotype in organisms ranging from microbes to humans, yet little is known about how natural selection acts on genome arrangement. Pathogenic bacteria such as Yersinia pestis, which causes bubonic and pneumonic plague, often exhibit a high degree of genomic rearrangement. The recent availability of several Yersinia genomes offers an unprecedented opportunity to study the evolution of genome structure and arrangement. We introduce a set of statistical methods to study patterns of rearrangement in circular chromosomes and apply them to the Yersinia. We constructed a multiple alignment of eight Yersinia genomes using Mauve software to identify 78 conserved segments that are internally free from genome rearrangement. Based on the alignment, we applied Bayesian statistical methods to infer the phylogenetic inversion history of Yersinia. The sampling of genome arrangement reconstructions contains seven parsimonious tree topologies, each having different histories of 79 inversions. Topologies with a greater number of inversions also exist, but were sampled less frequently. The inversion phylogenies agree with results suggested by SNP patterns. We then analyzed reconstructed inversion histories to identify patterns of rearrangement. We confirm an over-representation of "symmetric inversions"-inversions with endpoints that are equally distant from the origin of chromosomal replication. Ancestral genome arrangements demonstrate moderate preference for replichore balance in Yersinia. We found that all inversions are shorter than expected under a neutral model, whereas inversions acting within a single replichore are much shorter than expected. We also found evidence for a canonical configuration of the origin and terminus of replication. Finally, breakpoint reuse analysis reveals that inversions with endpoints proximal to the origin of DNA replication are nearly three times more frequent. Our findings represent the first characterization of genome arrangement evolution in a bacterial population evolving outside laboratory conditions. Insight into the process of genomic rearrangement may further the understanding of pathogen population dynamics and selection on the architecture of circular bacterial chromosomes.

Show MeSH
Related in: MedlinePlus