Computational Chemical and Structural Biology
Membrane proteins play a key role in a plethora of physiological processes and are thus popular drug targets. Even though significant advances have been made in structure determination methods, such as X-ray crystallography, NMR spectroscopy, and cryoEM microscopy, membrane proteins still make up only a small fraction of publicly available structures. Furthermore, few computational methods for de novo folding of membrane proteins have been presented. One potential alternative means of structure elucidation is to combine computational methods with experimental data. Hirst and others showed that this approach could be successfully applied to soluble proteins in 2011 (link to RosettaEPR paper). In this work, we present RosettaTMH, a novel algorithm for de novo membrane protein structure determination. We also show that distance data from EPR spectroscopy can be used in conjunction with RosettaTMH. RosettaTMH, with and without experimental restraints, performs better than the published MembraneAbinitio folding method and is capable of achieving native-like protein topologies in most cases, including rhodopsin.