Seminar Tomas Brazdil – Pushing the limits in automated NMR structure determination
December 13 @ 15 h 00 min
Pushing the limits in automated NMR structure determination
Tomas Brazdil (Masaryk University)
Nuclear Magnetic Resonance (NMR) spectroscopy is a technique for determining the structural and dynamic properties of molecules. NMR spectroscopy generates complex data whose analysis is a laborious task and the road to NMR structure is generally a long one. Our starting point is the 4D-CHAINS algorithm that performs a fully automated assignment of chemical shifts based on 4D TOCSY and 4D NOESY spectra. We propose to perform the assignment task using solely the 4D NOESY spectrum. Since Rosetta software can deliver accurate structures from a sparse network of distance constraints extracted from the NOESY spectrum, our approach promises to crack the atomic structures of large proteins in a single step. To solve the assignment problem, we utilize both machine learning as well as elementary graph theoretic methods. Machine learning techniques are used to train models for the amino acid and atom type prediction from a given set of chemical shifts. To overcome relative scarcity of 4D-NOESY data, we employ large amounts of chemical shifts deposited in Biological Magnetic Resonance Bank by human experts. The trained models are subsequently incorporated into an algorithm that assigns the corresponding sets of chemical shifts to amino acid residues of a given protein. In my talk, I will describe the core pipeline, explain the main difficulties and demonstrate the functionality of the current version which is still under development.