Titolo della tesi: Modeling the Spectroscopic Properties of Complex Systems of Biological Interest
The aim of this thesis was the development and application of novel theoretical and computational methods to model the photochemical properties of systems of biological interest in solution. This was achieved by using classical Molecular Dynamics (MD) and the Molecular Dynamics-Perturbed Matrix Method (MD-PMM), a mixed quantum/classical approach. New methodologies were implemented in the MD-PMM framework, including the calculation of the magnetic properties and the Circular Dichroism of molecules in solution, a simplified and cost-effective approach to model the vibrational overlaps, and a general approach to model the kinetics of slow processes where the proper sampling of the reactive events is not feasible. Moreover, an open-source program written in Python3 was developed to run MD-PMM simulations and to serve as the basis to implement existing and new methodologies to make them easily accessible. These approaches were applied to challenging and meaningful molecular systems to test their accuracy and gain important chemical insight. The spectroscopic behavior of Indole, the model for the sidechain of the amino acid Tryptophan (Trp), was reproduced in solution, including UV-Vis absorption and emission along an isobar. A model for the overall relaxation mechanism of excited Indole was also proposed. These results could serve as the starting point for understanding the complex fluorescence properties of Trp. The UV-Vis absorption properties of the anti-cancer drug Doxorubicin (DX) were calculated in different environments. The complex shape of its signal and its changes due to the polarity of the solvent were interpreted by identifying the contributions arising from two conformers of DX. The complexation of DX on polystyrene sulfonate chains was described using classical MD, providing a structural interpretation of the experimental data. Finally, the Circular Dichroism (CD) spectrum of L-Alanine in solution was reproduced by explicitly accounting for the most relevant degrees of freedom, providing the basis to expand the implemented methodology to larger systems such as peptides.