Error Tolerant NMR Backbone Resonance Assignment and Automated Structure Generation

摘要

Error tolerant backbone resonance assignment is the cornerstone of the NMR structure determinationnprocess. Although a variety of assignment approaches have been developed, none works sufficientlynwell on noisy fully automatically picked peaks to enable the subsequent automatic structure determinationnsteps.We have designed an integer linear programming (ILP) based assignment system (IPASS)nthat has enabled fully automatic protein structure determination for four test proteins. IPASS employsnprobabilistic spin system typing based on chemical shifts and secondary structure predictions. Furthermore,nIPASS extracts connectivity information from the inter-residue information and the (automaticallynpicked) 15N-edited NOESY peaks which are then used to fix reliable fragments. When appliednto automatically picked peaks for real proteins, IPASS achieves an average precision and recall ofn82% and 63%, respectively. In contrast, the next best method, MARS, achieves an average precisionnand recall of 77% and 36%, respectively. The assignments generated by IPASS are then fed into ournprotein structure calculation system, FALCON-NMR, to determine the 3D structures without humannintervention. The final models have backbone RMSDs of 1.25 ˚ A, 0.88 ˚ A, 1.49 ˚ A, and 0.67˚A to the referencennative structures for proteins TM1112, CASKIN, VRAR, and HACS1, respectively. The webnserver is publicly available at http://monod.uwaterloo.camr/ipass.