Bert Bogaerts

Multi-locus sequence typing (MLST) is a typing method to differentiate bacteria based on the sequence of several housekeeping genes. Identifiers are assigned to unique allele sequences for each locus in the scheme, and the combination of these identifiers defines a genetic profile. Its reproducibility and portability across laboratories have made it a staple typing method, widely used for epidemiological and evolutionary analyses. Although MLST was traditionally based on Sanger sequencing, the method is still widely used in the era of whole-genome sequencing (WGS). Moreover, WGS has made core-genome MLST (cgMLST) possible, which scale up MLST to hundreds or thousands of loci across the genome. Conventional and cgMLST schemes are publicly available on various platforms such as PubMLST.org, BIGSdb Institut Pasteur, cgMLST.org, and EnteroBase. However, the available software for (offline) cgMLST allele calling is often not flexible to accommodate schemes from diverse sources and/or lacks the computational scalability required to efficiently process larger schemes. In this study, we present Minimap2-inferred Sequence Typing (MiST), a rapid and flexible cgMLST allele caller that is low-resource intensive and can easily accommodate schemes from different sources. We benchmarked the tool against other available MLST and cgMLST calling software using WGS data from five different species and schemes collected from various sources. We demonstrate that MiST can accurately identify alleles while requiring substantially fewer computational resources than existing allele callers. MiST can help to make cgMLST analysis more accessible for integration into local bioinformatics workflows. MiST is available as an open-source Python package under the GPLv3 license at https://github.com/BioinformaticsPlatformWIV-ISP/MiST , and can be installed via Pip or Conda.

Core-genome multi-locus sequence typing (cgMLST) is a well-established and standardized method for genomics-based cluster detection and phylogenomic analysis of bacteria. The reduced error rate of Oxford Nanopore Technologies (ONT) R10 sequencing has prompted many laboratories to explore incorporating this technology into their activities. However, conflicting reports exist on the performance of ONT R10 sequencing for cgMLST analysis. This study evaluates the suitability of ONT R10 data for cgMLST allele calling and cluster detection for bacterial outbreak investigation. ONT and Illumina sequencing data were generated for 24 Neisseria meningitidis and 24 Salmonella enterica isolates. For ONT, both the rapid barcoding kit (RBK) and the rapid PCR barcoding kit (RPB) were used. The percentage of loci called in the ONT-only assemblies was very high for both species. However, the proportion of mismatched alleles to the hybrid assemblies was substantially higher for the Neisseria ONT-only assemblies with the RBK kit, resulting in incorrect cluster assignments. The large majority of these mismatched alleles were due to incorrect base calls at methylated positions, which did not affect the ONT data generated using the RPB kit or any of the Salmonella ONT-only assemblies. In conclusion, ONT R10 sequencing shows great potential as a viable method for cgMLST analysis, but methylation-related errors can affect performance for certain species and strains. When properly corrected for, ONT R10 had the same performance for cgMLST analysis as Illumina, and both could be used interchangeably. These results support the integration of ONT R10 sequencing into routine public health and clinical workflows. IMPORTANCE: This study evaluates the suitability of Oxford Nanopore Technologies R10 sequencing for core-genome multi-locus sequence typing (cgMLST), a widely used method in (clinical) outbreak investigation and bacterial strain tracking. We have sequenced 24 Neisseria meningitidis and 24 Salmonella enterica strains, including confirmed outbreak cases, using Illumina and ONT R10 sequencing to evaluate the performance for cgMLST analysis. We used a PCR-based and native barcoding protocol for the ONT sequencing, which enabled us to demonstrate a substantial species-dependent impact of methylation-related errors on the performance. However, we demonstrate that when these errors are properly addressed, ONT R10 can be used for accurate cgMLST-based clustering, including integration with strains sequenced using Illumina. Our findings support the use of ONT R10 as an alternative to Illumina sequencing for cgMLST analysis in routine public health practice.

Background: The Joint Programming Initiative on Antimicrobial Resistance (JPIAMR) networks ‘Seq4AMR’ and ‘B2B2B AMR Dx’ were established to promote collaboration between microbial whole genome sequencing (WGS) and antimicrobial resistance (AMR) stakeholders. A key topic discussed was the frequent variability in results obtained between different microbial WGS-related AMR gene prediction workflows. Further, comparative benchmarking studies are difficult to perform due to differences in AMR gene prediction accuracy and a lack of agreement in the naming of AMR genes (semantic conformity) for the results obtained. To illustrate this problem, and as a capacity-building exercise to encourage stakeholder involvement, a comparative Galaxy-based BenchAMRking platform was developed and validated using datasets from bacterial species with PCR-verified AMR gene presence or absence information from abritAMR. Results: The Galaxy-based BenchAMRking platform (https://erasmusmc-bioinformatics.github.io/benchAMRking/) specifically focusses on the steps involved in identifying AMR genes from raw reads and sequence assemblies. The platform currently comprises four well-characterised and published workflows that have previously been used to identify AMR genes using WGS data from several different bacterial species. These four workflows, which include the ISO certified abritAMR workflow, make use of different computational tools (or tool versions), and interrogate different AMR gene sequence databases. By utilising their own data, users can investigate potential AMR gene-calling problems associated with their own in silico workflows/protocols, with a potential use case outlined in this publication. Conclusions: BenchAMRking is a Galaxy-based comparison platform where users can access, visualise, and explore some of the major discrepancies associated with AMR gene prediction from microbial WGS data.

Infections due to antimicrobial-resistant bacteria have become a major threat to global health. Some patients may carry resistant bacteria in their gut microbiota. Specific risk factors may trigger the conversion of these carriages into infections in hospitalized patients. Preventively eradicating these carriages has been postulated as a promising preventive intervention. However, previous attempts at such eradication using oral antibiotics or probiotics have led to discouraging results. Phage therapy, the therapeutic use of bacteriophage viruses, might represent a worthy alternative in this context. Taking inspiration from this clinical challenge, we built Gut-On-A-Chip (GOAC) models, which are tridimensional cell culture models mimicking a simplified gut section. These were used to better understand bacterial dynamics under phage pressure using two relevant species: Pseudomonas aeruginosa and Escherichia coli. Model mucus secretion was documented by ELISA assays. Bacterial dynamics assays were performed in GOAC triplicates monitored for 72 h under numerous conditions, such as pre-, per-, or post-bacterial timing of phage introduction, punctual versus continuous phage administration, and phage expression of mucus-binding properties. The potential genomic basis of bacterial phage resistance acquired in the model was investigated by variant sequencing. The bacterial “escape growth” rates under phage pressure were compared to static in vitro conditions. Our results suggest that there is specific bacterial prosperity in this model compared to other in vitro conditions. In E. coli assays, the introduction of a phage harboring unique mucus-binding properties could not shift this balance of power, contradicting previous findings in an in vivo mouse model and highlighting the key differences between these models. Genomic modifications were correlated with bacterial phage resistance acquisition in some but not all instances, suggesting that alternate ways are needed to evade phage predation, which warrants further investigation.

Whole-genome sequencing has become the method of choice for bacterial outbreak investigation, with most clinical and public health laboratories currently routinely using short-read Illumina sequencing. Recently, long-read Oxford Nanopore Technologies (ONT) sequencing has gained prominence and may offer advantages over short-read sequencing, particularly with the recent introduction of the R10 chemistry, which promises much lower error rates than the R9 chemistry. However, limited information is available on its performance for bacterial single-nucleotide polymorphism (SNP)-based outbreak investigation. We present an open-source workflow, Prokaryotic Awesome variant Calling Utility (PACU) ( https://github.com/BioinformaticsPlatformWIV-ISP/PACU ), for constructing SNP phylogenies using Illumina and/or ONT R9/R10 sequencing data. The workflow was evaluated using outbreak data sets of Shiga toxin-producing Escherichia coli and Listeria monocytogenes by comparing ONT R9 and R10 with Illumina data. The performance of each sequencing technology was evaluated not only separately but also by integrating samples sequenced by different technologies/chemistries into the same phylogenomic analysis. Additionally, the minimum sequencing time required to obtain accurate phylogenetic results using nanopore sequencing was evaluated. PACU allowed accurate identification of outbreak clusters for both species using all technologies/chemistries, but ONT R9 results deviated slightly more from the Illumina results. ONT R10 results showed trends very similar to Illumina, and we found that integrating data sets sequenced by either Illumina or ONT R10 for different isolates into the same analysis produced stable and highly accurate phylogenomic results. The resulting phylogenies for these two outbreaks stabilized after ~20 hours of sequencing for ONT R9 and ~8 hours for ONT R10. This study provides a proof of concept for using ONT R10, either in isolation or in combination with Illumina, for rapid and accurate bacterial SNP-based outbreak investigation.

Methicillin-resistant Staphylococcus aureus (MRSA) constitutes a serious public health concern, with a considerable impact on patients’ health, and substantial healthcare costs. In this study, patients and healthcare workers (HCWs) from six public hospitals in Benin were screened for MRSA. Strains were identified as MRSA using conventional microbiological methods in Benin, and confirmed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in Belgium. Whole-genome sequencing (WGS) was used on the confirmed MRSA isolates, to characterize their genomic content and study their relatedness. Amongst the 305 isolates (304 wound swabs and 61 nasal swabs) that were collected from patients and HCWs, we detected 32 and 15 cases of MRSA, respectively. From this collection, 27 high-quality WGS datasets were obtained, which carried numerous genes and mutations associated with antimicrobial resistance. The mecA gene was detected in all the sequenced isolates. These isolates were assigned to five sequence types (STs), with ST8 (55.56%, n = 15/27), ST152 (18.52%, n = 5/27), and ST121 (18.52%, n = 5/27) being the most common. These 27 isolates carried multiple virulence genes, including the genes encoding the Panton–Valentine leukocidin toxin (48.15%, n = 13/27), and the tst gene (29.63%, n = 8/27), associated with toxic shock syndrome. This study highlights the need to implement a multimodal strategy for reducing the risk of the cross-transmission of MRSA in hospitals.

Methicillin-resistant Staphylococcus aureus (MRSA) constitutes a serious public health concern with a considerable impact on patients' health and substantial healthcare costs. In this study, patients, and healthcare workers (HCWs)from six public hospitals in Benin were screened for MRSA. Strains were identified as MRSA using conventional microbiological methods in Benin and confirmed using Matrix-Assisted Laser Desorption/Ionization Time of Flight (MALDI-TOF) mass spectrometry in Belgium. Whole genome sequencing (WGS) was used on confirmed MRSA isolates to characterize their genomic content and study their relatedness. In total,304 and 61 samples were collected from patients and HCWs, respectively. The mecA gene was detected in all isolates. The isolates were assigned to five sequence types (STs), with ST8 (55.6%, n=15), ST152 (18.52%, n=5), and ST121 (18.52%, n=5) being the most common. All isolates carried multiple virulence genes, including genes encoding the Panton-Valentine leukocidin toxin (48.15%, n=13) and the test gene (29.63%, n=8) associated with toxic shock syndrome. The presence of strains carrying numerous genomic features associated with antibiotic resistance and virulence in HCWs is a major public health concern. This study highlights the need to implement a multimodal strategy for reducing the risk of cross-transmission of MRSA in hospitals.

Genetically modified microorganisms (GMM) are frequently employed for manufacturing microbial fermentation products such as food enzymes or vitamins. Although the fermentation product is required to be pure, GMM contaminations have repeatedly been reported in numerous commercial microbial fermentation produce types, leading to several rapid alerts at the European level. The aim of this study was to investigate the added value of shotgun metagenomic high-throughput sequencing to confirm and extend the results of classical analysis methods for the genomic characterization of unauthorized GMM. By combining short- and long-read metagenomic sequencing, two transgenic constructs were characterized, with insertions of alpha-amylase genes originating from B. amyloliquefaciens and B. licheniformis, respectively, and a transgenic construct with a protease gene insertion originating from B. velezensis, which were all present in all four investigated samples. Additionally, the samples were contaminated with up to three unculturable Bacillus strains, carrying genetic modifications that may hamper their ability to sporulate. Moreover, several samples contained viable Bacillus strains. Altogether these contaminations constitute a considerable load of antimicrobial resistance genes, that may represent a potential public health risk. In conclusion, our study showcases the added value of metagenomics to investigate the quality and safety of complex commercial microbial fermentation products.