Investigating a Salmonella Typhi outbreak with Solu Platform

Salmonella Typhi is a bacterial pathogen that causes typhoid fever, a serious illness affecting millions of people worldwide. In this case study, we will explore the use of the Solu platform to investigate a Salmonella Typhi outbreak and analyze the genomic data associated with it.

Introduction

Salmonella Typhi outbreaks are significant public health challenges. Quick and accurate outbreak investigation methods are important for identifying the source, and coming up with effective interventions.

Traditional methods of outbreak investigation can be time-consuming and labor-intensive. New genomic sequencing technologies and advanced bioinformatics tools, like Solu, can rapidly analyze and interpret genomic data, providing valuable insights into outbreak dynamics.

This case study focuses on a specific outbreak of Salmonella Typhi, demonstrating how the Solu platform helps in understanding genomic data to for outbreak investigation.

Data sources

For this case study, we used data from a research paper titled "The rapid emergence of Salmonella Typhi with decreased ciprofloxacin susceptibility following an increase in ciprofloxacin prescriptions in Blantyre, Malawi", by Ashton and colleagues (2023).

The authors used Illumina HiSeq X ten for sequencing the isolates.

Link to the research paper

Analyzing the genomes in Solu platform

File upload

We start the process by uploading 20 genomes to Solu platform in Fasta format (assemblies). The upload took two minutes, and all analyses finished within 10 minutes, highlighting the platform's ability to process batches of genomic data.

Species, subspecies, serovar and strain

The platform does automatic species identification, and for S.enterica also subspecies and serovars. We quickly see that all but one sample belong to MLST 1 of Salmonella Typhi. This may indicate that several samples are sharing a common source of infection. We will later look at this in more detail, but for sample BZD6YF, we can quite confidently conclude that it is different from the other samples.

Quality analysis

The platform runs a number of quality assurance checks to ensure the accuracy and reliability of the genomic data. This step is crucial in ensuring the validity of subsequent analyses and interpretations.

The genome sizes are about 4.74Mbp, which is in the acceptable range.

The assembly statistics look to be OK, but the samples seem to have modest coverage. A coverage of >30x would be preferred for. This impacts some of the results, but does not seem alarming as the assembly looks otherwise all right.

‍

‍

For one sample, we also compared raw reads to the assembly. We uploaded raw reads to the Solu platform and waited for the platform to perform a de novo assembly and read quality analysis. The resulting summary shows that the sequencing quality has been good, with no large problems detected.

AMR genes

By switching to the “Resistome” tab, we see a high-level picture of resistance patterns in this dataset. AMR is a significant concern in outbreaks, as it can affect treatment options and the threat level of the outbreak.

Immediately, we see a few interesting findings

• The MLST 2 isolate (BZD6YF) does not have AMR genes, but other isolates are very homogenous in their resistance profile
• All MLST 1 isolates have the ESBL blaTEM-1 gene
• There was some variance in gyrA and gyrB mutations (probably explaining the reduced ciprofloxacin susceptibility in the clinical outcomes)

The individual sample page shows a more detailed antimicrobial resistance profile for the samples.  In addition to ESBL and ciprofloxacin (shown as quinolone in the table), the samples have chloramphenicol, streptomycin, sulfonamide and trimethorphin resistance genes. Interestingly, the samples also have some antiseptic resistance properties (by having the qacE delta 1 gene). This information could explain some of the MIC values, and inform the strategies to manage the outbreak.

Additional genomic profiling

Furthermore, the platform analyzed the presence of some virulence genes.

All isolates have the cdtB, iroB, iroC, and sinH virulence genes. The 19 MLST 1 isolates have the merC, merP, merR, and merT mercury resistance genes.

Phylogenetic tree and clustering

In addition to analyzing individual samples, the Solu platform also generates a phylogenetic tree based on pairwise SNP distances. This tree visualizes the genetic relationships between the different Salmonella Typhi samples, allowing us to identify clusters and potential transmission routes.

This tree confirms the above finding - one isolate (MLST 2) is very different from the others, but the other isolates appear very similar, which indicates that they belong to the same outbreak and have shared ancestors. In fact, all 19 of the samples are within 35 SNP.

‍

The platform automatically clusters genetically similar samples. Clustering analysis helps to find out potential sources of contamination.

The platform identified 3 clusters:

• St06 (15 samples within 23 SNP)
• St010 (3 samples within 7 SNP)
• St011 (2 samples within 5 SNP)

Here are the clusters overlaid on the tree. These clusters help us see the probable chains of transmission.

We also see that the clusters are very close to one another, meaning that the clusters are all part of the same outbreak.

Conclusions

Using this platform, we were able to quickly upload and analyze a large dataset of genomes, enabling us to identify important characteristics of the outbreak. The platform's ability to generate a phylogenetic tree and perform clustering analysis helped us understand the genetic relationships between different samples and identify potential sources of transmission. Additionally, the analysis of antimicrobial resistance genes and virulence genes provided insights into the bacteria's resistance patterns and pathogenicity. With the information obtained from the Solu platform, we can now make informed decisions regarding intervention strategies and treatment options for the outbreak.

‍