Abstract
With the decreasing cost of next-generation sequencing, whole-genome sequence-based bacterial genome comparisons are expected to become a mainstream process in the public health domain. Extended multilocus sequence typing (MLST) methods are becoming increasingly popular for use in comparing bacterial genetic relatedness in epidemiological investigations. Several extended MLST schemes based on biological signatures have been reported. Among them, whole-genome MLST (wgMLST) has gradually become one of the most widely used approaches for bacterial strain typing. In addition to using bacterial typing, many researchers aim to identify differences in the genes of compared strains. Because these differences might provide insights into critical bacterial functions, such as virulence and pathogenicity, researchers usually study these genes that differ between strains. Hence, we designed a web service tool based on wgMLST-constructed tree topology coupled with the feature selection method to create the “canonical wgMLST (cano-wgMLST) tree.” The genes that differ between strains are shown at each split of the tree, thereby directly providing information for performing a comparative genomic analysis for each strain pair. We demonstrated that this web service tool could be operated efficiently on two datasets consisting of 22 Campylobacter jejuni isolates and 59 S. Heidelberg isolates, respectively. We imposed this tool on a designated web server, cano-wgMLST_BacCompare, to enable users to create a wgMLST tree and canonical wgMLST tree automatically from their uploaded bacterial genomes for not only epidemiological investigation but also comparative genomic analysis. Additionally, detailed information on how to use this service is provided. The cano-wgMLST_BacCompare is available at http://baccompare.imst.nsysu.edu.tw.
Highlights
The origin of the multilocus sequence typing (MLST) approach can be traced back to 1998, when it was proposed by Maiden et al (1998) as a method for overcoming the typing data exchange problem
Neisseria meningitidis was the first species used to test the MLST approach, the approach has been proven to be applicable to many bacterial species, and the information of at least 100 species has been included in MLST allele databases (Jolley and Maiden, 2010; Jolley et al, 2018)
MLST-based approaches are useful to public health researchers, the costs of MLST experiments were high in the Sanger sequencing era
Summary
The origin of the multilocus sequence typing (MLST) approach can be traced back to 1998, when it was proposed by Maiden et al (1998) as a method for overcoming the typing data exchange problem. With the reduction in cost of nextgeneration sequencing, many studies have attempted to evaluate whole-genome MLST schemes for comparing the genomes of several common pathogenic bacteria, such as Salmonella Enteritidis (Pearce et al, 2018), Listeria monocytogenes (Chen et al, 2016), Vibrio parahaemolyticus (Gonzalez-Escalona et al, 2017), Enterococcus faecium (de Been et al, 2015), and Campylobacter jejuni (Cody et al, 2017) These studies have tended to extend the MLST scheme from housekeeping genes to whole-genome genes (i.e., whole-genome MLST and wgMLST) and demonstrated that wgMLST has favorable discriminatory power for distinguishing highly closely related strains. The genetic relatedness tree and heatmap profile indicating different genes for each split on the basis of the final created scheme are displayed on the result page
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
