Double-stranded coarse grained model for DNA: applications to supercoiling and denaturation

DNA supercoiling is the name given to the under or overwinding of the two strands of a DNA double helix. It is of great interest because it is relevant in several crucial biological processes. However, the principles governing its dynamics and its precise role under different circumstances remain elusive. Despite advances in single molecule experimental techniques, measuring supercoiling dynamics persist a challenge; this is where computer simulations are useful. In this thesis, I first introduce a single-nucleotide resolution coarse-grained computational model of DNA, that faithfully reproduces the geometry of the double-stranded helix and also part of its elastic behaviour. The dynamic of the system is implemented using a molecular dynamics scheme, and the results obtained are interpreted through methods of equilibrium and non-equilibrium statistical mechanics. I then employ this model to specifically study DNA supercoiling. This phenomenon, although topological in nature, is extremely important for the survival of cells because it has a deep impact on the regulation of gene expression, the compaction of DNA inside the cell and DNA replication. In particular, this work finds its motivations in: (i) an experimentally unresolved problem about the effect of supercoiling on DNA melting; (ii) the dynamics of supercoiling under physiological conditions during transcription; and (iii) the relation between supercoiling and DNA-binding proteins. Given that these phenomena may be relevant in vivo, they have recently received a a great deal of attention. However, until now, no computational model existed to study these kind of process. The techniques used here have been successful in providing insight into the key elements in the system. This would have been impossible before by using, for example, 1D models. The major achievement of this work is the quantitative characterisation of the role played by DNA supercoiling in a range of situations that are commonly found in vivo