Table 3.

Analysis of the Restrained-MD Generated Structures of d(TTGGCCAA)₂·Chr₂·Co(II)

NMR Pseudocontact Shift Restraints
total no. of measured paramagnetic shifts	256
no. of pseudocontact shifts used in refinement	234
no. of pseudocontact shifts for DNA	168
no. of pseudocontact shifts for drug	66
range of pseudocontact shifts	−45.42 to +42.50
NMR Distance Restraints
total number of distance restraints	788
intraresidue distance restraints	492
interresidue distance restraints	296
Empirical Restraints
H-bonding restraints	20
Co – O distance restraints	4
Structural Statistics
	SHFTa,b	NOEa,c	SHFT_NOEa,d
rmsd of NOE violations (Å)	0.34 ± 0.01	0.031 ± 0.005	0.036 ± 0.005
no. of NOE violations >0.3 Å	55 ± 2	0	0
rmsd of shift violations (ppm)	0.31 ± 0.01	7.93 ± 2.37	0.37 ± 0.04
no. of shift violations >expt+grad. erre	32 ± 4	227 ± 29	36 ± 5
Pairwise rmsd over all non-H atomsf
SHFTb,g	0.95 ± 0.20	1.99 ± 0.19	1.11 ± 0.17
NOEc,g	1.93 ± 0.25	2.05 ± 0.27	1.85 ± 0.25
SHFT+NOEd,g	1.11 ± 0.17	1.90 ± 0.20	1.01 ± 0.04
rmsd from ideal geometry
bond length (Å)	0.017 ± 0.0	0.016 ± 0.0	0.018 ± 0.001
bond angle (deg)	2.94 ± 0.04	3.15 ± 0.31	3.74 ± 0.01
improper angle (deg)	4.00 ± 0.17	3.65 ± 0.34	3.78 ± 0.30

^aThree families of structures; P (20 structures), B (20 structures), and A (10 structures).

^bThree structures obtained using pseudocontact shift data.

^cThree structures obtained using NOE data.

^dThree structures obtained using both pseudocontact shift and NOE data.

^eExperimental error is the sum of the line widths in the paramagnetic and diamagnetic spectra (tol in eq 2). Gradient. error is the shift caused by a 0.3 Å displacement along the steepest gradient (eq 3).

^fAverage of nine rmsd’s for comparison of structures obtained using different data sets, and of six rmsd’s for comparison of structures obtained with a given data set. Terminal base pairs are excluded.

^gThree minimized averaged structures for the three sets of restrained minimization, starting with P, B, and A (see text).