U.S. flag

An official website of the United States government

NIH: Fogarty International Center NIH: Fogarty International Center
Advancing Science for Global Health
Advancing Science for Global Health
Home > International Epidemiology & Population Studies > Bacterial Molecular Evolution Print

Bacterial Molecular Evolution

Illustration that shows the phylogenetic tree for the cluster of E. coli containing the recurrent outbreak isolates

For several bacterial species, available whole-genome DNA sequences number in the tens or hundreds of thousands. Such datasets allow reconstruction of evolutionary history that includes thousands of sequence changes that have occurred fairly recently. The inferred changes provide information about both the rates of various types of mutation and the forces of natural selection. The study of recent sequence changes, made possible by the availability of sequences of very closely-related isolates, provides an unusual window into natural selection. Purifying selection, which eliminates sufficiently harmful mutations in the long run, has had little time to act, so that only highly deleterious changes are appreciably affected. Furthermore, sequence changes that confer an advantage under rare conditions, but are disfavored and eliminated in the long run, can be identified.


Selection-Driven Gene Inactivation

For a protein-coding gene to persist, selection must favor a functional protein in the long term. Nevertheless, premature stop codons in some genes appear to be positively selected under some circumstances. The identities of affected genes shed light on both the evolutionary phenomenon and the genes themselves.

Determinants of Position-Specific Mutation Rates

Nucleotide positions in a genome exhibit different rates of mutation. Previous work on this project characterized the high C→T mutation rate at 5-methyl C residues in bacteria, and used this to infer methylation patterns. Other work revealed that N4 methylation of C by a particular Type III restriction-modification system is associated with extremely high C→A mutation rate. Additional hypermutation phenomena are under study.

Short-Term Effects of Selection on Coding Sequences

The evolution of coding sequences and their protein products is most commonly studied by comparison of sequences from different species, with divergence times of millions of years. Comparison of sequences from tight genetic clusters of bacteria, with divergence times of years to decades, provides a very different perspective on protein evolution.

Evolution of Pathogen Characteristics

Genetic differences within a bacterial species can have phenotypic consequences that are relevant to disease. Reconstruction of genetic changes that have occurred recently can elucidate the evolution of disease-causing and other important traits on a relatively short time scale (years or decades).




  • Joshua L. Cherry
  • Nídia S. Trovão