Investigating the mutational consequences of transcription-damage interactions through computational modelling

• martin.taylor@ed.ac.uk
• michael.nicholson@ed.ac.uk

Investigating the mutational consequences of transcription-damage interactions through computational modelling.

Cellular DNA needs to be read accurately and efficiently during transcription. Yet every day, each genome is subjected to hundreds of thousands of chemical modifications, known as DNA damage, which can impede the transcriptional machinery. Damage-induced transcription dysregulation is observed across organisms and is hypothesised to be a primary driver of physiological ageing (1). However, it’s currently unclear how frequently the transcriptional machinery is impeded, what eventually happens to the machinery if it is impeded, and how this depends on the type of damage. These gaps preclude a quantitative understanding of the physiological consequences of DNA damage. 

The outcomes of transcription-damage interactions influence DNA mutation patterns. Exploiting this relationship, our group recently used mouse tumour mutations to measure how frequently different interaction outcomes occur, for a specific type of DNA damage (2, 3). How these measurements vary for other damage types or in different experimental systems is unclear.

To address these knowledge gaps, this PhD project aims to:

• Characterise transcription-damage associated mutation patterns in a variety of systems including in vitro experiments where cells have been exposed to specific damaging agents, and cancer genomes which record the outcomes of transcription-damage interactions along somatic lineages. This will require development and application of a range of bioinformatics skills.
• Develop mathematical models describing the evolution of mutation patterns resulting from transcription-damage through somatic cellular lineages. We anticipate using stochastic simulations and tools from Markov processes.
• Compare the model predictions to the mutation data to measure interaction outcomes. We anticipate using likelihood or Bayesian inference methods. 

The ideal candidate for this interdisciplinary PhD will have strong mathematical/computational skills, e.g. a degree in Mathematics, Physics, or another discipline with substantial mathematical training and experience with coding. Extensive biological knowledge is not initially required, but the candidate should be motivated to learn, to answer biological research questions, and to interact with both theoretical and biology researchers.

The PhD programme will provide training in molecular biology, computational genomics, mathematical modelling, and statistical inference, through supervisor guidance and/or formal courses. The candidate will also gain skills in communicating with diverse academic teams and broader professional skills through EastBio training opportunities. 

References:

1. Lans et al. Nat Rev Mol Cell Biol (https://doi.org/10.1038/s41580-019-0169-4 )
2. Nicholson et al, PNAS, 2024 (https://doi.org/10.1073/pnas.2403871121)
3. Anderson et al, Nature, 2024 (https://doi.org/10.1038/s41586-024-07490-1)