A Machine Learning Model for Food Source Attribution of Listeria monocytogenes.

Despite its low morbidity, listeriosis has a high mortality rate due to the severity of its clinical manifestations. The source of human listeriosis is often unclear. In this study, we investigate the ability of machine learning to predict the food source from which clinical Listeria monocytogenes isolates originated. Four machine learning classification algorithms were trained on core genome multilocus sequence typing data of 1212 L. monocytogenes isolates from various food sources. The average accuracies of random forest, support vector machine radial kernel, stochastic gradient boosting, and logit boost were found to be 0.72, 0.61, 0.7, and 0.73, respectively. Logit boost showed the best performance and was used in model testing on 154 L. monocytogenes clinical isolates. The model attributed 17.5 % of human clinical cases to dairy, 32.5% to fruits, 14.3% to leafy greens, 9.7% to meat, 4.6% to poultry, and 18.8% to vegetables. The final model also provided us with genetic features that were predictive of specific sources. Thus, this combination of genomic data and machine learning-based models can greatly enhance our ability to track L. monocytogenes from different food sources.

Exploring the predictive capability of advanced machine learning in identifying severe disease phenotype in Salmonella enterica.

The past few years have seen a significant increase in availability of whole genome sequencing information, allowing for its incorporation in predictive modeling for foodborne pathogens to account for inter- and intra-species differences in their virulence. However, this is hindered by the inability of traditional statistical methods to analyze such large amounts of data compared to the number of observations/isolates. In this study, we have explored the applicability of machine learning (ML) models to predict the disease outcome, while identifying features that exert a significant effect on the prediction. This study was conducted on Salmonella enterica, a major foodborne pathogen with considerable inter- and intra-serovar variation. WGS of isolates obtained from various sources (i.e., human, chicken, and swine) were used as input in four machine learning models (logistic regression with ridge, random forest, support vector machine, and AdaBoost) to classify isolates based on disease severity (extraintestinal vs. gastrointestinal) in the host. The predictive performances of all models were tested with and without Elastic Net regularization to combat dimensionality issues. Elastic Net-regularized logistic regression model showed the best area under the receiver operating characteristic curve (AUC-ROC; 0.86) and outcome prediction accuracy (0.76). Additionally, genes coding for transcriptional regulation, acidic, oxidative, and anaerobic stress response, and antibiotic resistance were found to be significant predictors of disease severity. These genes, which were significantly associated with each outcome, could possibly be input in amended, gene-expression-specific predictive models to estimate virulence pattern-specific effect of Salmonella and other foodborne pathogens on human health.

Transcriptome and Comparative Genomics Analyses Reveal New Functional Insights on Key Determinants of Pathogenesis and Interbacterial Competition in Pectobacterium and Dickeya spp.

Soft-rot Enterobacteriaceae (SRE), typified by Pectobacterium and Dickeya genera, are phytopathogenic bacteria inflicting soft-rot disease in crops worldwide. By combining genomic information from 100 SRE with whole-transcriptome data sets, we identified novel genomic and transcriptional associations among key pathogenicity themes in this group. Comparative genomics revealed solid linkage between the type I secretion system (T1SS) and the carotovoricin bacteriophage (Ctv) conserved in 96.7% of Pectobacterium genomes. Moreover, their coactivation during infection indicates a novel functional association involving T1SS and Ctv. Another bacteriophage-borne genomic region, mostly confined to less than 10% of Pectobacterium strains, was found, presumably comprising a novel lineage-specific prophage in the genus. We also detected the transcriptional coregulation of a previously predicted toxin/immunity pair (WHH and SMI1_KNR4 families), along with the type VI secretion system (T6SS), which includes hcp and/or vgrG genes, suggesting a role in disease development as T6SS-dependent effectors. Further, we showed that another predicted T6SS-dependent endonuclease (AHH family) exhibited toxicity in ectopic expression assays, indicating antibacterial activity. Additionally, we report the striking conservation of the group 4 capsule (GFC) cluster in 100 SRE strains which consistently features adjacently conserved serotype-specific gene arrays comprising a previously unknown organization in GFC clusters. Also, extensive sequence variations found in gfcA orthologs suggest a serotype-specific role in the GfcABCD machinery.IMPORTANCE Despite the considerable loss inflicted on important crops yearly by Pectobacterium and Dickeya diseases, investigations on key virulence and interbacterial competition assets relying on extensive comparative genomics are still surprisingly lacking for these genera. Such approaches become more powerful over time, underpinned by the growing amount of genomic information in public databases. In particular, our findings point to new functional associations among well-known genomic themes enabling alternative means of neutralizing SRE diseases through disruption of pivotal virulence programs. By elucidating novel transcriptional and genomic associations, this study adds valuable information on virulence candidates that could be decisive in molecular applications in the near future. The utilization of 100 genomes of Pectobacterium and Dickeya strains in this study is unprecedented for comparative analyses in these taxa, and it provides novel insights on the biology of economically important plant pathogens.

Transcriptome Profiling Reveals the EanI/R Quorum Sensing Regulon in Pantoea Ananatis LMG 2665T.

Pantoea ananatis LMG 2665T synthesizes and utilizes acyl homoserine lactones (AHLs) for signalling. The complete set of genes regulated by the EanI/R quorum sensing (QS) system in this strain is still not fully known. In this study, RNA-sequencing (RNA-seq) was used to identify the EanI/R regulon in LMG 2665T. Pairwise comparisons of LMG 2665T in the absence of AHLs (Optical density (OD)600 = 0.2) and in the presence of AHLs (OD600 = 0.5) were performed. Additionally, pairwise comparisons of LMG 2665T and its QS mutant at OD600 = 0.5 were undertaken. In total, 608 genes were differentially expressed between LMG 2665T at OD600 = 0.5 versus the same strain at OD600 = 0.2 and 701 genes were differentially expressed between LMG 2665T versus its QS mutant at OD600 = 0.5. A total of 196 genes were commonly differentially expressed between the two approaches. These constituted approximately 4.5% of the whole transcriptome under the experimental conditions used in this study. The RNA-seq data was validated by reverse transcriptase quantitative polymerase chain reaction (RT-qPCR). Genes found to be regulated by EanI/R QS were those coding for redox sensing, metabolism, flagella formation, flagella dependent motility, cell adhesion, biofilm formation, regulators, transport, chemotaxis, methyl accepting proteins, membrane proteins, cell wall synthesis, stress response and a large number of hypothetical proteins. The results of this study give insight into the genes that are regulated by the EanI/R system in LMG 2665T. Functional characterization of the QS regulated genes in LMG 2665T could assist in the formulation of control strategies for this plant pathogen.