Multilevel genome typing (MGT) enables genomic characterization of bacterial isolates and their relationships using multiple multilocus sequence typing (MLST)-like schemes within the one system. In MGT, smaller schemes provide lower resolution and larger schemes provide higher resolution. Thus, relationships between isolates assigned MGT can be studied at multiple levels of resolutions. The MGT designation for an isolate is stable, and assignment is faster compared to existing approaches. MGT has been developed and successfully applied to describe three pathogens, namely Salmonella enterica serovars Typhimurium and Enteritidis, and Vibrio cholerae.
Here, we present a publicly accessible web service called MGTdb (https://mgtdb.unsw.edu.au), which enables the assignment, storage, retrieval, and analysis of isolates characterized using MGT. The MGTdb web service enables a user to upload isolates as sequence reads or extracted alleles. Additionally, any newly sequenced isolates deposited in NCBI SRA are also regularly retrieved, processed, assigned MGT and made publicly available. Interactive visualisation tools are built-in to assist analysis, along with capabilities to download all publicly available isolates and assignments, as well as, to export files for use with external software.
We demonstrate the features of MGTdb, and its usability through three case studies, 1) via interactive visualisations built into MGTdb and streamlined export of data for Microreact – using V. chloerae, we demonstrate its usefulness to study the global epidemiology of a species, 2) via national-level-reports consisting of static graphs - using S. Typhimurium in UK, we demonstrate its usefulness for national surveillance, and 3) via filter functionality and streamlined export of data for GrapeTree – using S. Typhimurium, we demonstrate its usefulness to study outbreaks. Thus, epidemiological and microbiological investigations such as those determining if isolates belong to previous, ongoing or new outbreaks, pandemics or epidemics can be easily performed, at multiple levels of resolution in MGTdb.