Oral Presentation The Prato Conference on the Pathogenesis of Bacterial Infections of Animals 2016

Genotyping, Epidemiological Associations and Prediction of the Phenotype of Escherichia coli Strains using Superphy (#11)

Matthew D Whiteside 1 , Chad R Laing 1 , Peter Kruczkiewicz 1 , Ed Taboada 1 , Victor Gannon 1
  1. Public Health Agency of Canada, Lethbridge, ALBERTA, Canada

While most members of the species Escherichia coli are considered intestinal commensals, others have been assigned to a variety of “pathogroups” based on their association with either enteric or extra-intestinal infections. The identificationof characteristics of these disease-associated groups has been central to the diagnosis, treatment and prevention of E. coli-associated illness. Advances in nucleotide sequencing technology accompanied by decreased costs offer unprecedented amount of genomic data to workers on this pathogen. However, how can this information best be leveraged to meet the needs of diverse user groups such as epidemiologists, evolutionary biologists, ecologists and clinicians? The bioinformatics community has responded enthusiastically to this challenge and developed a variety of tools for specific analytical tasks. We initially developed a software platform called “Panseq” for the identification of core and accessory genomes and identification of differences in the genomes of groups of bacterial species. However, we saw the need for a more integrated set of “biotools” linked to genome databases to facilitate routine implementation of whole genome sequence analysis into research, clinical and reference laboratories.The “Superphy” platform (https://lfz.corefacility.ca/superphy/) provides an up to date E. coli genome database to which users can upload their own sequences and obtain inventories of virulence attributes and antimicrobial resistance determinants for strains of interest and performin silico subtyping and determine phylogenies of these determinants. There is also a module which allows the identification of biomarkers i.e. genetic markers significantly associated with metadata categories such as host or specific diseases. Powerful inferences can also be made with respect to the long-term evolutionary origins and geographical dispersal of subtypes as well as short-term changes in genomes that occur during outbreaks or within individual hosts. Specific examples will be given to demonstrate the functionalities of the modules of this software platform.