Large-scale motions and specificity are two aspects of conformational change in biomolecules. This work explores both perspectives and in each case proposes a method that coarse-grains a difficult problem into more tractable fragments.

The normal mode basis set comes out as a natural representation of the collective motions of the system. First, I quantitatively assess here the accuracy and limitations of this simple representation. Second, I propose a molecular dynamics simulation trajectory in which the time integration is limited to the low frequency normal modes of the system. The objective of this approach is the extension of the simulation trajectory.

Regarding specificity of biomolecules, an atomistic fragment-based approach breaks up the intricate energy landscape between a substrate and a ligand into simpler components of the force field, giving insights into biological phenomenology. I take the ribosome exit tunnel as an example of allosteric system, where the nascent chain is the ligand that interacts with the complex cavity of the tunnel. I computationally study the features of the tunnel that have a role in the regulation of translation.