Table 1. Results of rosetta fold prediction and filtering with unassigned NMR data.

Protein			NMR data			Ranking of rosetta models using NMR data
PDB ID code	Fold type	No. of residues	CS	NOE	RDC	Correlation coefficient*	Best-scoring cluster rmsd, ņ	Correctly assigned backbone atoms, %‡
1b4c	α	92	472	830	218	-0.73	4.58	3.1
1cmz	α	128	1,042	1,221	104	-0.72	6.67	1.4
1gbl	αβ	56	471	612	283	-0.90	4.46	6.6
1ghh	αβ	81	771	760	530	-0.86	4.81	10.9
1khm	αβ	88	735	1,128	206	-0.87	4.45	4.1
1ubi	αβ	76	803	1,240	628	-0.91	3.43	13.5
2ezm	β	101	767	1,005	327	-0.81	6.23	3.2
2ezxA	α	89	808	729	245	-0.85	6.29	2.7

^*Correlation coefficient between the rmsd of the nine protein models to the native structure and the average score of the top five assignments for each of the models

^†rmsd of the best-scoring cluster to the native structure.

^‡Average percentage of correctly assigned backbone atoms for the best-scoring cluster center.