New software developed at CEITEC automates NMR structural studies

Researchers from CEITEC Masaryk University developed a streamlined strategy designed to meet the key objectives of NMR structure determination, that is minimal data collection, least human intervention, and applicability to large proteins.

Thomas Evangelidis from Kostas Tripsianes research group developed the computer software that provides highly accurate and near-complete NMR resonance assignments by greatly reducing the number of required experiments and human time spent. The breakthrough for automation was the use of 4D spectra that offer higher resolution and allow NMR studies of larger structures compared to traditional 3D NMR approaches. 4D-CHAINS was successful in assigning automatically a protein composed of 248 amino acids.

NMR is a structural biology tool to study the structure of proteins in solution. Standard NMR methods rely on recording a large number of multi-dimensional spectra to establish chemical shift assignments, which are then used to determine the structure of proteins.

The main barrier currently is that obtaining NMR assignments at sufficient levels of completeness and correctness requires several weeks of measurement time and several months of effort by a trained expert, in other words the road to NMR structure is generally a long one. New software “4D-CHAINS” is an algorithm that provides fully automated assignments of NMR chemical shifts from only two complementary NMR spectra.

“The whole idea was based on using a minimal set of 4D spectra to expedite the acquisition time, reduce the NMR experimental cost, minimize the human input required, but most importantly provide higher precision in data analysis,” said Kostas Tripsianes the leader of research team.

Algorithm 4D-CHAINS enables automated assignments of chemical shifts at 95% completeness with 1.5% error rate. In combination with autoNOE-Rosetta, 4D-CHAINS provides a robust approach leveraging a highly automated process to obtain reliable structures in a matter of days.

“Our results on a blind benchmark set of targets show a tremendous reduction in human effort and NMR spectrometer time needed to obtain data-driven, high-resolution structures,” said Tripsianes. The full process takes approximately three weeks to execute for a typical protein sample, including the time needed for NMR data acquisition, and requires minimum supervision.

Masaryk university