

Pathogenic Escherichia coli is a major cause of foodborne illness worldwide and includes strains capable of causing severe disease. To establish a genome-informed framework for foodborne outbreak surveillance, we analyzed 1,029 E. coli isolates from clinical, food, livestock, and environmental sources using whole-genome sequencing. Pathogenic isolates obtained from human clinical cases or linked to documented outbreaks were classified as epidemiologically defined high-risk (EpiHR), whereas the remaining pathogenic isolates were classified as non-EpiHR. Virulence-associated genomic features were extracted using a bioinformatics pipeline, and four machine learning (ML) algorithms, including gradient boosting machine, random forest (RF), and support vector machines with linear and radial basis function kernels, were evaluated. Among them, the RF model showed the best performance, achieving an area under the curve (AUC) of 0.98 and accuracy of 0.93 in 10-fold cross-validation. Additional leave-one-group-out validation showed retained discrimination across held-out sequence types and serotypes, although performance was reduced when isolates were grouped by isolation source. Evaluation using an independent test dataset of 1,908 publicly available pathogenic E. coli genomes showed an AUC of 0.97 and a sensitivity of 0.98. Feature importance analysis using Shapley additive explanations identified influential predictive features, including traT, etpB, and enterotoxin-associated genes. A reduced 10-feature model achieved an AUC of 0.79 in the independent test dataset, supporting its exploratory use for future simplified screening approaches. These results indicate that genome-based ML provides a sensitive framework for surveillance-oriented prioritization of EpiHR pathogenic E. coli isolates, with model predictions interpreted together with epidemiological information.
Actinobacillus pleuropneumoniae (APP) is the etiological agent of porcine pleuropneumoniae (PP), a high contagious respiratory disease with significant impact on the swine industry in both clinically and economically. Despite of the several attempts to control APP, the emergence of novel serotypes and antimicrobial resistance (AMR) strains highlights the importance of monitoring the genetic characteristics of APP at single nucleotide level. Despite the importance of genomic surveillance of APP to develop effective control strategies, genetic information on the recent Korean isolates of APP is not available at whole genome level. Therefore, in this study, six APP strains were isolated from porcine lungs with characteristic lesions of PP from 2022 to 2024. And their whole genomic sequences, serotypes, virulence factors, and AMR traits were investigated using combined short- and long-read sequencing methods. In silico PCR serotyping identified the isolates as serotype 1, 7, and 15, while one isolate was non-typeable. Multiple AMR genes including Hinf_PBP3_BLA, Ecol_EFTu_PLV, tet(B), tet(O), tetR, sul2, aph(3'')-Ib, aph(6)-Id, and aph(3')-Ia were detected. Also, these genes were located with adjacent to mobile genetic elements, suggesting the possibility of horizontal gene transfer. Phylogenetic comparison with 40 global APP complete genomes, presented that Korean isolates were closely related with China and Switzerland strains. This study provides the whole genome sequences based genetic characterization on the recent Korean isolates of APP, and this study emphasizes that continuous monitoring of APP genomic variation to support effective control of porcine pleuropneumoniae.
With the advent of whole-genome sequencing, opportunities to investigate the population structure, transmission patterns, antimicrobial resistance profiles, and virulence determinants of Streptococcus pneumoniae at high resolution have been increasingly expanding. Consequently, a user-friendly bioinformatics tool is needed to automate the analysis of Streptococcus pneumoniae whole-genome sequencing data, summarize clinically relevant genomic features, and further guide treatment options. Here, we developed PneusPage, a web-based tool that integrates functions for species prediction, molecular typing, drug resistance determination, and data visualization of Streptococcus pneumoniae. To evaluate the performance of PneusPage, we analyzed 80 pneumococcal genomes with different serotypes from the Global Pneumococcal Sequencing Project and compared the results with those from another platform, PathogenWatch. We observed a high concordance between the two platforms in terms of serotypes (100% concordance rate), multilocus sequence typing (100% concordance rate), penicillin-binding protein typing (88.8% concordance rate), and the Global Pneumococcal Sequencing Clusters (98.8% concordance rate). In addition, PneusPage offers integrated analysis functions for the detection of virulence and mobile genetic elements that are not provided by previous platforms. By automating the analysis pipeline, PneusPage makes whole-genome sequencing data more accessible to non-specialist users, including microbiologists, epidemiologists, and clinicians, thereby enhancing the utility of whole-genome sequencing in both research and clinical settings. PneusPage is available at
Citations
Citations
Citations