Hi, Charles: By using the following operation I can make a perfect match between the calculated CS value with output from XPLOR. CS_xplor = ( CS - sigma_iso ) * Da / 10. So how did the Da value come into play? And why is the scaling factor is Da / 10 instead of Da? Our XPLOR-NIH script is based on Tian/Marassi's script. In the script we used DaScale = -20, and a Da value of 10.735. I don't understand what these parameters mean. Can you give a clue? Thanks. Jian
________________________________________ From: Charles Schwieters [[email protected]] Sent: Wednesday, June 26, 2013 9:15 PM To: DAI, JIAN Cc: xplor-nih at nmr.cit.nih.gov Subject: Re: [Xplor-nih] Back-calculate chemical shift anisotropy Hello Jian-- I'm assuming you're using create_CSAPot from the csaPotTools module, and that the atom selections are in this order: C N HN. Also, if you are manually setting the components of sigma (rather than updating csaPotTools.csaData), the values should probably be scaled up by 1000. > When calculating the tensors, what definition was used. > For 15N chemical shift, by reading the source code, it seems as they > are calculated as follows: > e3 = r_NH X r_NC e_z = unit vec along (q_H-q_N) x (q_C-q_N) e_y = unit vec along (q_H-q_N) x e_z e_x = e_y x e_z if gamma=0, then the rotation operation is (e1,e2,e3 ) = (e_x,e_y,e_z ) * Rz(beta) where (e1,e2,e3) represents a matrix composed of columns of vectors, and Rz(beta) is a clockwise rotation matrix about the z axis. I'm traveling at the moment, so I can't make a precise comparison with your expression at this time. best regards-- Charles
