As molecular machines, proteins not only often adopt well-defined structures but also typically must possess internal dynamics including particularly flexible regions and motion along specific modes. Such flexible regions and large-scale internal motions, however, create challenges for protein structure determination. In particular, Nuclear Magnetic Resonance (NMR)-based structure calculations may fail to converge in flexible loops or yield a unique inter-domain orientation. A lack of convergence in atomic coordinates within an NMR structure complicates the application of NMR-derived structural data to problems such as Molecular Replacement (MR). While knowledge of protein flexibility is important for understanding protein function as well as in applying NMR-derived structural data, the prediction of protein flexibility, the relation of protein flexibility to lack of convergence in NMR-based structure calculations and how best to exclude ill-defined regions of NMR-derived structure in applying those structures in MR are all areas of active research. This talk presents my research group’s progress in predicting protein flexibility from sequence data. In particular, we have developed a method for predicting Lipari-Szabo order parameters solely from sequence data that produces results on par with methods requiring high resolution protein structures. This talk also will present a comparison between coordinate uncertainty in NMR-derived structures, Crystallographic B-factors and coordinate variances from Molecular Dynamics simulations demonstrating that lack of convergence in NMR-based structure calculations indeed indicates protein flexibility. Additionally, this talk will report our progress in applying FindCore2 to the problem of using NMR-derived structures for MR.

Protein flexibility and NMR: How can we predict and deal with it?

251st National Meeting (2016)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.