On modeling field dependence of paramagnetic NMR observables and other neglected aspects in structural biology

Integration of different experimental techniques is becoming the new paradigm for Structural Biology for understanding biomolecular structure and dynamics. By combining experimental data from diverse techniques with computational modeling, Integrative Structural Biology (ISB) aims to achieve a comprehensive and accurate understanding of biomolecular functions. Most of the current ISB methodologies rely upon the use of empirical or classical computational models, limiting the accuracy of the final result. Quantum chemical (QC) methods hold the potential to bridge this gap. In order to make substantial progress toward quantum-level refinement, it is imperative to develop computational protocols that can accurately reproduce the experimental data. My intention throughout my doctoral work was to move a step in this direction, focusing on one of the most used constraints in structural biology, i.e. the paramagnetic contribution to the chemical shift. NMR spectroscopy, a mainstay of ISB, is highly sensitive to the structural environment of biomolecules and is particularly effective for the study of paramagnetic systems, where the hyperfine interaction between unpaired electrons and nuclei provides valuable structural details. The availability of ultrahigh magnetic fields has added a further complication in the interpretation of paramagnetic NMR spectra, not only in terms of the increased influence of relaxation, but because a theoretical framework for describing the impact of magnetic field on paramagnetic observables was missing. This thesis addresses this challenge by developing a computational framework for simulating NMR parameters in open-shell systems analyzed at high field strengths. The first part of the study thus focuses on computational protocols for calculating hyperfine shifts in paramagnetic systems across varying magnetic fields, with an emphasis on small inorganic complexes as a model system. These protocols are designed to be broadly applicable, and ultimately extendable to more complex bioinorganic molecules. This thesis addresses a further problem in the use of paramagnetic NMR observables, providing algorithms and techniques for processing and analyzing the spectra, specifically considering the effects and limitations associated with high-field acquisitions. Finally, given the ensemble nature of most biological systems and the limitations of single-structure modeling, this thesis explores a geometrical approach for ensemble data analysis. This approach is particularly relevant for intrinsically disordered systems, where the vast conformational space sampled cannot be accurately represented by a single structural model. By incorporating an ensemble-based perspective, this thesis aims to advance the precision of structural analysis and lay the groundwork for quantum-level refinement in integrative structural biology, ultimately moving toward a robust framework for high-accuracy biomolecular simulations.