We describe a fully automated two-step approach to NMR resonance assignments and de novo structure determination of large proteins without deuteration. First, the CHAINS algorithm uses a combination of two 4D spectra –a TOCSY and a NOESY, providing residue type and distance information, respectively– to perform resonance assignments of aliphatic sidechain atoms in addition to the backbone amide groups, thus allowing near-complete coverage of 1H probes along the protein sequence. The automated assignments are then imported to iterative NOE cross-peak assignment and structure calculations in Rosetta, to generate high-resolution NMR structures of proteins. The novel approach goes beyond the current state-of-the-art methods in NMR structure determination in three aspects: NMR experimental time, automation in data analysis, and size of proteins. The GAMU E project will complement the research findings on the current protein benchmark with experimental data on an even larger target to be included in the manuscript for submission to Nature Methods journal.

Publikace