Efficient inference of recombination hot regions in bacterial genomes

Koji Yahara, Xavier Didelot, M. Azim Ansari, Samuel K. Sheppard, Daniel Falush

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

In eukaryotes, detailed surveys of recombination rates have shown variation at multiple genomic scales and the presence of "hotspots" of highly elevated recombination. In bacteria, studies of recombination rate variation are less developed, in part because there are few analysis methods that take into account the clonal context within which bacterial evolution occurs. Here, we focus in particular on identifying "hot regions" of the genome where DNA is transferred frequently between isolates. We present a computationally efficient algorithm based on the recently developed "chromosome painting" algorithm, which characterizes patterns of haplotype sharing across a genome. We compare the average genome wide painting, which principally reflects clonal descent, with the painting for each site which additionally reflects the specific deviations at the site due to recombination. Using simulated data, we show that hot regions have consistently higher deviations from the genome wide average than normal regions. We applied our approach to previously analyzed Escherichia coli genomes and revealed that the new method is highly correlated with the number of recombination events affecting each site inferred by ClonalOrigin, a method that is only applicable to small numbers of genomes. Furthermore, we analyzed recombination hot regions in Campylobacter jejuni by using 200 genomes. We identified three recombination hot regions, which are enriched for genes related to membrane proteins. Our approach and its implementation, which is downloadable from https://github.com/bioprojects/ orderedPainting, will help to develop a new phase of population genomic studies of recombination in prokaryotes.

Original languageEnglish
Pages (from-to)1593-1605
Number of pages13
JournalMolecular Biology and Evolution
Volume31
Issue number6
Early online date27 Feb 2014
DOIs
Publication statusPublished - 1 Jun 2014

Keywords

  • chromosome painting
  • homologous recombination
  • population genomics
  • recombination hot region
  • recombination hotspot
  • selection

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Molecular Biology
  • Genetics

Cite this

Efficient inference of recombination hot regions in bacterial genomes. / Yahara, Koji; Didelot, Xavier; Ansari, M. Azim; Sheppard, Samuel K.; Falush, Daniel.

In: Molecular Biology and Evolution, Vol. 31, No. 6, 01.06.2014, p. 1593-1605.

Research output: Contribution to journalArticle

Yahara, Koji ; Didelot, Xavier ; Ansari, M. Azim ; Sheppard, Samuel K. ; Falush, Daniel. / Efficient inference of recombination hot regions in bacterial genomes. In: Molecular Biology and Evolution. 2014 ; Vol. 31, No. 6. pp. 1593-1605.
@article{354a81da852645c9977a91b782ae0848,
title = "Efficient inference of recombination hot regions in bacterial genomes",
abstract = "In eukaryotes, detailed surveys of recombination rates have shown variation at multiple genomic scales and the presence of {"}hotspots{"} of highly elevated recombination. In bacteria, studies of recombination rate variation are less developed, in part because there are few analysis methods that take into account the clonal context within which bacterial evolution occurs. Here, we focus in particular on identifying {"}hot regions{"} of the genome where DNA is transferred frequently between isolates. We present a computationally efficient algorithm based on the recently developed {"}chromosome painting{"} algorithm, which characterizes patterns of haplotype sharing across a genome. We compare the average genome wide painting, which principally reflects clonal descent, with the painting for each site which additionally reflects the specific deviations at the site due to recombination. Using simulated data, we show that hot regions have consistently higher deviations from the genome wide average than normal regions. We applied our approach to previously analyzed Escherichia coli genomes and revealed that the new method is highly correlated with the number of recombination events affecting each site inferred by ClonalOrigin, a method that is only applicable to small numbers of genomes. Furthermore, we analyzed recombination hot regions in Campylobacter jejuni by using 200 genomes. We identified three recombination hot regions, which are enriched for genes related to membrane proteins. Our approach and its implementation, which is downloadable from https://github.com/bioprojects/ orderedPainting, will help to develop a new phase of population genomic studies of recombination in prokaryotes.",
keywords = "chromosome painting, homologous recombination, population genomics, recombination hot region, recombination hotspot, selection",
author = "Koji Yahara and Xavier Didelot and Ansari, {M. Azim} and Sheppard, {Samuel K.} and Daniel Falush",
year = "2014",
month = "6",
day = "1",
doi = "10.1093/molbev/msu082",
language = "English",
volume = "31",
pages = "1593--1605",
journal = "Molecular Biology and Evolution",
issn = "0737-4038",
publisher = "Oxford University Press",
number = "6",

}

TY - JOUR

T1 - Efficient inference of recombination hot regions in bacterial genomes

AU - Yahara, Koji

AU - Didelot, Xavier

AU - Ansari, M. Azim

AU - Sheppard, Samuel K.

AU - Falush, Daniel

PY - 2014/6/1

Y1 - 2014/6/1

N2 - In eukaryotes, detailed surveys of recombination rates have shown variation at multiple genomic scales and the presence of "hotspots" of highly elevated recombination. In bacteria, studies of recombination rate variation are less developed, in part because there are few analysis methods that take into account the clonal context within which bacterial evolution occurs. Here, we focus in particular on identifying "hot regions" of the genome where DNA is transferred frequently between isolates. We present a computationally efficient algorithm based on the recently developed "chromosome painting" algorithm, which characterizes patterns of haplotype sharing across a genome. We compare the average genome wide painting, which principally reflects clonal descent, with the painting for each site which additionally reflects the specific deviations at the site due to recombination. Using simulated data, we show that hot regions have consistently higher deviations from the genome wide average than normal regions. We applied our approach to previously analyzed Escherichia coli genomes and revealed that the new method is highly correlated with the number of recombination events affecting each site inferred by ClonalOrigin, a method that is only applicable to small numbers of genomes. Furthermore, we analyzed recombination hot regions in Campylobacter jejuni by using 200 genomes. We identified three recombination hot regions, which are enriched for genes related to membrane proteins. Our approach and its implementation, which is downloadable from https://github.com/bioprojects/ orderedPainting, will help to develop a new phase of population genomic studies of recombination in prokaryotes.

AB - In eukaryotes, detailed surveys of recombination rates have shown variation at multiple genomic scales and the presence of "hotspots" of highly elevated recombination. In bacteria, studies of recombination rate variation are less developed, in part because there are few analysis methods that take into account the clonal context within which bacterial evolution occurs. Here, we focus in particular on identifying "hot regions" of the genome where DNA is transferred frequently between isolates. We present a computationally efficient algorithm based on the recently developed "chromosome painting" algorithm, which characterizes patterns of haplotype sharing across a genome. We compare the average genome wide painting, which principally reflects clonal descent, with the painting for each site which additionally reflects the specific deviations at the site due to recombination. Using simulated data, we show that hot regions have consistently higher deviations from the genome wide average than normal regions. We applied our approach to previously analyzed Escherichia coli genomes and revealed that the new method is highly correlated with the number of recombination events affecting each site inferred by ClonalOrigin, a method that is only applicable to small numbers of genomes. Furthermore, we analyzed recombination hot regions in Campylobacter jejuni by using 200 genomes. We identified three recombination hot regions, which are enriched for genes related to membrane proteins. Our approach and its implementation, which is downloadable from https://github.com/bioprojects/ orderedPainting, will help to develop a new phase of population genomic studies of recombination in prokaryotes.

KW - chromosome painting

KW - homologous recombination

KW - population genomics

KW - recombination hot region

KW - recombination hotspot

KW - selection

U2 - 10.1093/molbev/msu082

DO - 10.1093/molbev/msu082

M3 - Article

VL - 31

SP - 1593

EP - 1605

JO - Molecular Biology and Evolution

JF - Molecular Biology and Evolution

SN - 0737-4038

IS - 6

ER -