Consistent Protein Structure Determination Using Sparse NMR Data

Abstract

Conventional NMR structure determination for proteins typically is labor-intensive and usually is challenging for larger proteins because only sparse NMR restraint data are generally available. Numerous alternative protein structure determination protocols have been developed to utilize inputs from a sparse set of experimental NMR data, comprising mainly backbone chemical shifts, residual dipolar couplings (RDCs) and sparse 1H–1H NOEs if available. These structure determination approaches directly exploit the powerful bioinformatics algorithms previously developed for sequence-based protein structure prediction and homology modeling, implemented with the essential structural information provided by a variety of sparse NMR data, and have been demonstrated for routinely generating accurate high-resolution full-atom structures for proteins with size up to ca. 40 kDa and with varying fold complexity and oligomeric states. This review aims to highlight the central concepts and important aspects of these sparse NMR data based protein structure determination protocols, represented by those consistent approaches developed within the CS-Rosetta framework.