Extended multi-locus sequence typing (eMLST) methods have become popular in the field of genomic epidemiology. Before eMLST methods can be applied in epidemiological investigations, the selection of a suitable scheme is critical. The core genome scheme (cgMLST) has become the most popular eMLST approach for strain typing in the epidemiological domain. In addition to strain typing, many public health researchers and clinical microbiologists wish to investigate which genes cause genetic differences between compared strains. Therefore, a tool that can be used to extract canonical genes with an eMLST scheme would be particularly useful. In this study, we present cano-eMLST, a well-designed program that applies a feature-selection methodology to create a canonical locus combination with discriminatory power by traversing a genetic relatedness tree based on a user-selected scheme. The cano-eMLST program is provided mainly to help infectious disease laboratory researchers identify potential factors related to bacterial pathogenesis. The core program (tree-traversing approach) of cano-eMLST is implemented in Perl and Python. All the necessary dependencies and environmental settings are provided in the encapsulated version (VirtualBox or VMware) and self-installation version (all use source code and libraries).
Keywords: core-genome multi-locus sequence typing (cgMLST); feature-selection; molecular typing; next-generation sequencing (NGS).