NMR peak assignment

Keywords: Nuclear Magnetic Resonance, Residues, Assignment problem
Project partners: Marta Stefanska, Dr. Adam Mazur, Dr. Thomas Müntener, Elsa Stern, Prof. Dr. Sebastian Hiller
CeDA collaborator: Konstantinos Ntemos

Nuclear Magnetic Resonance (NMR) is a physical phenomenon in which atomic nuclei with non-zero spin, when placed in a strong external magnetic field, absorb and re-emit electromagnetic radiation at characteristic resonance frequencies. These frequencies depend on the local electronic and structural environment of each nucleus, making NMR a powerful method for studying molecular structure, dynamics, and composition.

In this work, we focus on the residue-level NMR peak assignment problem for proteins. This consists of assigning each observed peak in an NMR spectrum to the protein residue that generated it. More specifically, we investigate whether residue-level assignments can be inferred from 2D ¹H–¹⁵N HSQC spectra. HSQC is a relatively inexpensive and widely used experiment, but the assignment problem is challenging in this setting because the spectrum provides limited information, typically one backbone amide ¹H–¹⁵N cross-peak per non-proline residue.

Some of the research avenues to be explored are the following:

1. Explore the NMR peak values synthetic generation algorithms.
2. Discrepancy between synthetic and experimental data.
3. Algorithmic solution to solve the problem with minimal information (e.g., only N15-H1 peak values).

Develop an algorithmic solution for automated NMR peak assignment.