Prevalence and genetic diversity of human diarrheagenic Escherichia coli isolates by multilocus sequence typing

•This study facilitated an overall understanding of human diarrheagenic Escherichia coli (DEC) distributed worldwide.•Clonal complex (CC) 10 was the most widely distributed CC, comprising 276 isolates with 85 sequence types (STs).•Sequence variations were more abundant in fumC and gyrB than in the other five genes.•The population structure of five common DEC pathotypes was highly diversified.•Isolates with the same ST or CC were heterogeneous for different pathotypes. The population structure of human diarrheagenic Escherichia coli (DEC) isolates derived from worldwide collections remains undefined. A total of 1196 clinical isolates were obtained from a multilocus sequence typing (MLST) database. Genetic diversity analysis, MLST analysis, and phylogenetic analysis combined with different pathotypes were performed through a variety of calculation software applications. All isolates were categorized as one of 579 different sequence types (STs). The eBURST algorithm resolved these 579 STs into 27 clonal complexes (CCs), 37 concatemers, and 210 singletons, revealing a high level of genetic diversity in the population structure of DEC. CC10 was the most prevalent CC, comprising 276 (23.08%, 276/1196) isolates with 85 (14.68%, 85/579) STs widely distributed in 20 countries. The population structure of five common pathotypes was highly diversified, and isolates with the same ST or CC were heterogeneous for different pathotypes. Sequence variations were more abundant in fumC and gyrB than in the other five genes, and these exhibited the highest degree of nucleotide diversity (0.03886 and 0.03075, respectively) and the greatest number of polymorphic nucleotide sites (137 and 139, respectively). The dN/dS ratios of seven analyzed loci varied from 0.0083 (recA) to 0.0434 (purA), and the ratio for the concatenated sequence was 0.2518, revealing the effects of purifying selection on housekeeping genes during the evolutionary process. Significant allele linkage disequilibrium was detected when the standardized index of association (ISA) was calculated both for the entire collection of isolates (0.3174, p