Controlled vocabularies for microbial virulence factors

VFG-Chip: A high-throughput qPCR microarray for profiling virulence factor genes from the environment

As zoonotic pathogens are threatening public health globally, the virulence factor genes (VFGs) they carry underlie latent risk in the environment. However, profiling VFGs in the environment is still in its infancy due to lack of efficient and reliable quantification tools. Here, we developed a novel high-throughput qPCR (HT-qPCR) chip, termed as VFG-Chip, to comprehensively quantify the abundances of targeted VFGs in the environment. A total of 96 VFGs from four bacterial pathogens including Klebsiella pneumoniae, Acinetobacter baumannii, Escherichia coli, and Salmonella enterica were targeted by 120 primer pairs, which were involved in encoding five types of virulence factors (VFs) like toxin, adherence, secretion system, immune evasion/invasion, and iron uptake. The specificity of VFG-Chip was both verified computationally and experimentally, with high identity of amplicon sequencing and melting curves analysis proving its robust capability. The VFG-Chip also displayed high sensitivity (by plasmid serial dilution test) and amplification efficiency averaging 97.7%. We successfully applied the VFG-Chip to profile the distribution of VFGs along a wastewater treatment system with 69 VFGs detected in total. Overall, the VFG-Chip provides a robust tool for comprehensively quantifying VFGs in the environment, and thus provides novel information in assessing the health risks of zoonotic pathogens in the environment.

Suppressed inflammation in obese children induced by a high-fiber diet is associated with the attenuation of gut microbial virulence factor genes

In our previous study, a gut microbiota-targeted dietary intervention with a high-fiber diet improved the immune status of both genetically obese (Prader-Willi Syndrome, PWS) and simple obese (SO) children. However, PWS children had higher inflammation levels than SO children throughout the trial, the gut microbiota of the two cohorts was similar. As some virulence factors (VFs) produced by the gut microbiota play a role in triggering host inflammation, this study compared the characteristics and changes of gut microbial VF genes of the two cohorts before and after the intervention using a fecal metagenomic dataset. We found that in both cohorts, the high-fiber diet reduced the abundance of VF, and particularly pathogen-specific, genes. The composition of VF genes was also modulated, especially for offensive and defensive VF genes. Furthermore, genes belonging to invasion, T3SS (type III secretion system), and adherence classes were suppressed. Co-occurrence network analysis detected VF gene clusters closely related to host inflammation in each cohort. Though these cohort-specific clusters varied in VF gene combinations and cascade reactions affecting inflammation, they mainly contained VFs belonging to iron uptake, T3SS, and invasion classes. The PWS group had a lower abundance of VF genes before the trial, which suggested that other factors could also be responsible for the increased inflammation in this cohort. This study provides insight into the modulation of VF gene structure in the gut microbiota by a high-fiber diet, with respect to reduced inflammation in obese children, and differences in VF genes between these two cohorts.

Comparative metagenomic analysis of microbial taxonomic and functional variations in untreated surface and reclaimed waters used in irrigation applications

The use of irrigation water sourced from reclamation facilities and untreated surface water bodies may be a practical solution to attenuate the burden on diminishing groundwater aquifers. However, comprehensive microbial characterizations of these water sources are generally lacking, especially with regard to variations through time and across multiple water types. To address this knowledge gap we used a shotgun metagenomic approach to characterize the taxonomic and functional variations of microbial communities within two agricultural ponds, two freshwater creeks, two brackish rivers, and three water reclamation facilities located in the Mid-Atlantic, United States. Water samples (n = 24) were collected from all sites between October and November 2016, and filtered onto 0.2 μm membrane filters. Filters were then subjected to total DNA extraction and shotgun sequencing on the Illumina HiSeq platform. From these data, we found that Betaproteobacteria dominated the majority of freshwater sites, while Alphaproteobacteria were abundant at times in the brackish waters. One of these brackish sites was also host to a greater abundance of the bacterial genera Gimesia and Microcystis. Furthermore, predicted microbial features (e.g. antibiotic resistance genes (ARGs) and Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) arrays) varied based on specific site and sampling date. ARGs were found across samples, with the diversity and abundance highest in those from a reclamation facility and a wastewater-impacted freshwater creek. Additionally, we identified over 600 CRISPR arrays, containing ∼2600 unique spacers, suggestive of a diverse and often site-specific phage community. Overall, these results provide a better understanding of the complex microbial community in untreated surface and reclaimed waters, while highlighting possible environmental and human health impacts associated with their use in agriculture.

Predicting bacterial virulence factors - evaluation of machine learning and negative data strategies

Bacterial proteins dubbed virulence factors (VFs) are a highly diverse group of sequences, whose only obvious commonality is the very property of being, more or less directly, involved in virulence. It is therefore tempting to speculate whether their prediction, based on direct sequence similarity (seqsim) to known VFs, could be enhanced or even replaced by using machine-learning methods. Specifically, when trained on a large and diverse set of VFs, such may be able to detect putative, non-trivial characteristics shared by otherwise unrelated VF families and therefore better predict novel VFs with insignificant similarity to each individual family. We therefore first reassess the performance of dimer-based Support Vector Machines, as used in the widely used MP3 method, in light of seqsim-only and seqsim/dimer-hybrid classifiers. We then repeat the analysis with a novel, considerably more diverse data set, also addressing the important problem of negative data selection. Finally, we move on to the real-world use case of proteome-wide VF prediction, outlining different approaches to estimating specificity in this scenario. We find that direct seqsim is of unparalleled importance and therefore should always be exploited. Further, we observe strikingly low correlations between different feature and classifier types when ranking proteins by VF likeness. We therefore propose a 'best of each world' approach to prioritize proteins for experimental testing, focussing on the top predictions of each classifier. Further, classifiers for individual VF families should be developed.

A comparison of computational methods for identifying virulence factors

Bacterial pathogens continue to threaten public health worldwide today. Identification of bacterial virulence factors can help to find novel drug/vaccine targets against pathogenicity. It can also help to reveal the mechanisms of the related diseases at the molecular level. With the explosive growth in protein sequences generated in the postgenomic age, it is highly desired to develop computational methods for rapidly and effectively identifying virulence factors according to their sequence information alone. In this study, based on the protein-protein interaction networks from the STRING database, a novel network-based method was proposed for identifying the virulence factors in the proteomes of UPEC 536, UPEC CFT073, P. aeruginosa PAO1, L. pneumophila Philadelphia 1, C. jejuni NCTC 11168 and M. tuberculosis H37Rv. Evaluated on the same benchmark datasets derived from the aforementioned species, the identification accuracies achieved by the network-based method were around 0.9, significantly higher than those by the sequence-based methods such as BLAST, feature selection and VirulentPred. Further analysis showed that the functional associations such as the gene neighborhood and co-occurrence were the primary associations between these virulence factors in the STRING database. The high success rates indicate that the network-based method is quite promising. The novel approach holds high potential for identifying virulence factors in many other various organisms as well because it can be easily extended to identify the virulence factors in many other bacterial species, as long as the relevant significant statistical data are available for them.

Named entity recognition for bacterial Type IV secretion systems

Research on specialized biological systems is often hampered by a lack of consistent terminology, especially across species. In bacterial Type IV secretion systems genes within one set of orthologs may have over a dozen different names. Classifying research publications based on biological processes, cellular components, molecular functions, and microorganism species should improve the precision and recall of literature searches allowing researchers to keep up with the exponentially growing literature, through resources such as the Pathosystems Resource Integration Center (PATRIC, patricbrc.org). We developed named entity recognition (NER) tools for four entities related to Type IV secretion systems: 1) bacteria names, 2) biological processes, 3) molecular functions, and 4) cellular components. These four entities are important to pathogenesis and virulence research but have received less attention than other entities, e.g., genes and proteins. Based on an annotated corpus, large domain terminological resources, and machine learning techniques, we developed recognizers for these entities. High accuracy rates (>80%) are achieved for bacteria, biological processes, and molecular function. Contrastive experiments highlighted the effectiveness of alternate recognition strategies; results of term extraction on contrasting document sets demonstrated the utility of these classes for identifying T4SS-related documents.