Beyond vaccine research: biomarker discovery with long read RNA sequencing

Irina introduced herself as a postdoctoral researcher at the University of Oxford’s vaccine research group, and also introduced the study organism that was the subject of her talk, Salmonella typhi. Irina began by explaining that S. typhi is the causative agent of typhoid, and is responsible for approximately 100,000 deaths worldwide each year. The majority of these deaths are seen in low to middle income countries. Irina explained that S. typhi produces the typhoid toxin and that, as part of a human challenge study to understand gene expression during infection, they exposed 40 healthy individuals to either a wild type strain capable of producing the toxin, or an attenuated strain that was not. Interestingly, approximately 70% of participants in each group became sick, suggesting that the toxin itself was not required for disease manifestation and the development of typhoid fever. Furthermore, in a subset of people, the toxin-negative strain produced more severe clinical symptoms.

After outlying the experimental setup, Irina went on to talk about the use of nanopore sequencing to study differences in gene expression profiles between the two groups of individuals. Irina described how nanopore sequencing was a novel approach able to generate long, full-length transcripts, which may give the ability to produce more precise differential gene measurements and potentially resolve isoforms. Using whole blood collected from individuals from both challenged groups on day 0 and post-diagnosis on day 6–8, RNA was extracted and a polyA purification step was performed. Barcoded sequencing libraries were generated using the PCR-cDNA Sequencing Kit and paired samples per patient, i.e. one from day 0 and one post-diagnosis, were run together on single PromethION Flow Cells for 48 hours. This generated a median sequencing depth of 11,300,000 mapped reads per sample and a mean transcript length of 900 nt.

Irina then moved on to the data analysis section of her talk, and showed an ordination plot highlighting the fact that the dissimilarity between the transcriptomic profiles was smaller within each treatment group than between. This suggested large differences in the transcriptomic landscape between those exposed to a toxin producing bacterium and those exposed to a non-toxin producing variant. Showing a volcano plot to highlight the large number of up- and down- regulated genes identified as part of a differential expression analysis, Irina delved into more details of specific upregulated biochemical pathways. Genes belonging to the immune system response pathways were highly upregulated, as were those involved in interferon mediated pathways, specifically that of interferon gamma response. Irina suggested that this data could be used to identify potential biomarkers of infection.

Irina finished her talk with a teaser of the work the group is currently undertaking, that being studying vaccine response as part of the SARS-CoV-2 vaccine trials, something she will be able to share updates about in future. To summarise, Irina said that this was proof that long-read RNA sequencing with nanopore could be used in clinical research studies and could identify bacterial infection signatures. Furthermore, Irina described how ten million reads per sample gave sufficient resolution, and that this was clearly the future for her field of research.