iqbal-lab
commented
3 years ago It's good you're getting words down and finding your voice, but the abstract is too long. You just need to say what you're going to do. The intro is there to set the scene and explain why it is important.
Here's an example of what you could do
Genetic variation in bacterial genomes has a direct impact on their biology, and since they reproduce asexually, comparing genomes allows inference of relatedness. In the context of bacterial pathogens, this has important implications for global public health, empowering epidemiology and allowing prediction of antimicrobial drug resistance (amr).
A single reference genome is often used as a template to find variation, and this idea underlies the vast majority of bioinformatic approaches to studying genetic variation. I introduce this thesis by explaining why this approach is inadequate for bacteria, and describe a new approach to analyse bacterial genomes, based on a "pan-genome graph". In chapter 1, i describe algorithms for de novo assembly of genetic variation in the context of a pan-genome graph. The remaining chapters address an applied question: can nanopore sequence analysis of m tuberculosis genomes be used to provide high quality analysis for public health? To this end i collect data from south africa, madagascar and england. In chapter 2 i evaluate standard and graph-genome based methods for detecting potential outbreaks. In chapter 3 i address drug resistance prediction. in the final chapter i develop a tuberculosis-specific neural network model for base-calling nanopore data. Together this thesis addresses general methods for studying any bacterial species, and a specific question with potential impact for public health.
mbhall88
This needs to be submitted when requesting examiners along with the title and can't be changed prior to submission.