Detection of homologous recombination in closely related strains

Anastasia S. Kalinina, Alexandra L. Suvorikova, Vladimir G. Spokoiny, Mikhail S. Gelfand

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)


Detection of recombination events in a bacterial genome is both important from the evolutionary point of view, and of practical interest. Indeed, homologous recombination (HR) plays a major role in the exchange of antigenic determinants between strains. There exist statistical methods to detect recently recombined segments in whole-genome sequences that use a high local density of substitutions as a signal of HR events with a source outside considered strains. However, it is difficult to detect the HR events within a set of strains, which represent whole species diversity, due to a low number of substitutions in recombined segments and high level of diversity of strains. Here, we analyzed HR in 20 Escherichia coli (E. coli) strains to define what fraction of segments with a high substitution rate were introduced in a genome by HR. For detection of HR, we used the segmentation, performed by the adaptive weights smoothing (AWS) algorithm. It detects sharp changes in the structure of observed data analyzing only qualitative structural information. We validated the approach on simulated data, applied it to the analysis of E. coli strains, and determined the recombination rates between phylogroups.

Original languageEnglish
Article number1641001
JournalJournal of Bioinformatics and Computational Biology
Issue number2
Publication statusPublished - 1 Apr 2016
Externally publishedYes


  • AWS
  • change-point detector
  • Detection of homologous recombination
  • Escherichia coli


Dive into the research topics of 'Detection of homologous recombination in closely related strains'. Together they form a unique fingerprint.

Cite this