

This directory contains files derived from real NMR data for CASP COMMONS target UW_eng.

The Dihedral and Ambiguous Contact files have been revised to use the CASP residue numbering.  Specifically, these files reformatted with the new index - residue 1 was residue 18 in the original file. The purification tag is indexed from -17 to -1 before residue 1 (was numbered 1 to 17 in the original file).

The assessment will only be done for the CASP Target sequence, and it is not necessary to submit coordinates for the N-terminal 17 residue purification tag.

The protein sequence was provided by Brian Koepnick and David Baker at University of Washington.

The 13C,15N enriched protein sample was produced in the laboratories of Gaetano Montelione and Masayori Inouye at Rutgers.  The CASP COMMONS sample production project involved contributions by undergraduate students Briana, and Carlie Hanlon, and staff scientists Natalia Denissova, Keiko Inouye, and Project Leader Yojiro Ishida.

NMR data was collected by G.V.T Swapna, and analyzed by Gaohua Liu.  Data collection included standard triple-resonance NMR experiments for backbone and sidechain amide and methyl resonance assignments, as well as 3D N-NOESY and C-NOESY.  

The NMR analysis included only backbone HN, N, Ca, Cb, and C' resonances, sidechain amide H and N resonances, and sidechain H and C methyl resonances assignments. These classes of resonance assignments can be obtained most rapidly even for larger (15 - 60 kDa) perdeuterated proteins.  However, as this protein sample was not perdeuterated, NOESY peaks potentially arise from all H-atoms; the generation of ambiguous contacts from the NOESY spectra therefore includes a significant number of peaks for which the true assignment involves unassigned sidechain H atoms.   As a result, many of the ambiguous contacts are False Positives (i.e. none of the possible contacts listed is true).  This may be true of as much as 30% of the Ambiguous Contacts.

Ambiguous contact lists were generated from these resonance assignments and real NOESY peak lists using Cycle0 of the program ASDP (formally AutoStructure) [Y.J. Huang, R. Tejero, R. Powers, and G.T. Montelione, G.T. Proteins 62, 587-603 (2006)]

Dihedral Restraints were determined from these backbone resonance assignments using Talo_N [Y. Shen, and A. Bax, J. Biomol. NMR, 56, 227-241(2013)].

There are no RDCs or ECs available for this target.

This 'fast data collection' strategy can provide accurate structures, when combined with advanced modeling methods.

Following the Expiration Date of this real NMR data target N1008, we will release a second data set for the same protein, in which all backbone and sidechain resonance assignments are used in assigning NOESY peaks and generating the contact list.  This data set, typical of what can be routinely obtained for small (< 15 kDa) proteins, will have a much higher density of contacts, less ambiguity, and a much lower False Positive rate.

The high-accuracy reference structure, based on extensive sidechain resonance assignments and additional NMR data, will be provided after the second target Expiration Date.

G. T. Montelione
A. Rosato
H. Valafar
CASP NMR Assessors
