Theoretical advances in signal analysis of DEER spectroscopy

Abstract

Electron paramagnetic resonance (EPR) spectroscopy has been established as an indispensable tool for a wide range of scientific fields. Within the field of EPR, pulse methods, particularly those that measure dipolar couplings, are valuable for resolving protein structures. These applications have driven spectroscopists to address some key challenges in the methodology. This thesis discusses two challenges in the field of pulsed dipolar EPR spectroscopy. The first challenge involves improving the applicability of DEER spectroscopy by expanding the toolkit for data analysis. Due to the ill-conditioned nature of the problem, robust uncertainty quantification is a necessity. The work presented in this thesis provides fully quantified uncertainty of DEER data using Bayesian statistical methods. The second addresses the fundamental behavior of spins when placed in a highly polarized environment and identifies both analytically and experimentally how the generated signal deviates from what has been previously predicted.