| Summary: | This thesis explores the use of genomics to understand two organisms, Mycobacterium bovis and Escherichia coli.
M. bovis, the causative agent of tuberculosis in cattle, was isolated from a known wildlife host, badgers. The badgers in this study were collected as roadkill by a network of stakeholders and the study investigators. Over 700 badgers were collected, and approximately 640 useable carcasses were obtained. From these carcasses, ~12% were culture positive for M. bovis colonies, as confirmed through IS6110 PCR. Spoligotyping was performed on these isolates using a microarray. M. bovis colonies were sequenced and analysed in a variety of ways. Molecular spoligotyping was found to conform to In silico spoligotyping, but In silico typing was better able to resolve ambiguous types. Comparative analysis of the genomes found gaps in the genomes across several isolates. A control set of M. bovis reads from a badger were assembled using the same process to confirm that these regions were not an error in the method. Only one genomic gap was found not to be an artefact of the methods used. SNP analysis of the badger M. bovis sequences, combined with their geographical metadata, shows that there might be evidence of an additional transmission driver between badgers - due to the distances separating the identified clades. The small numbers of M. bovis isolates obtained in this study would likely not be enough to perform quantitative transmission dynamics analysis but highlights the difficulty in obtaining useful sequence from M. bovis in badgers, which is key to understanding multi-host transmission dynamics.
One isolate in this study M. bovis 867, was selected for sequencing using long-read sequencing technologies. Comparative analysis of the hybrid assembled M. bovis genome to the existing reference genomes found several SNPs distinguishing it from the reference. These SNPs were explored based on their predicted effects on protein translation and how this may phenotypically alter these isolates. However, phenotypic studies must be carried out to confirm these predictions. Overall, a method is presented for getting HMW, long-fragment DNA from M. bovis, from which long-read data can be generated. The data show the improvements to assemblies using a hybrid approach.
An additional study involving E. coli is presented as the final chapter of this thesis. This chapter investigates the improvements that can be made using hybrid assembly methods in microbial genomics. Twelve putative E. coli isolated from the feet of sheep on a farm were selected for hybrid assembly. From these 12 E. coli, whole contiguous plasmid sequences were resolved, including a 120kb megaplasmid carrying several AMR genes. The use of hybrid assembly methods also elucidated a Tn21-like transposon, a powerful driver of AMR not often seen in farm environments in the UK.
|
|---|