[Chimera-users] RMSD per residue NMR structure

Elaine Meng meng at cgl.ucsf.edu
Tue Jan 14 09:46:31 PST 2014

Hi Francesca,
The per-residue RMSD (from CA atoms only) can be calculated easily by associating all of the structures to a single sequence:

(1) open the NMR ensemble ensemble, then show the sequence for any one of the ensemble members, it doesn't matter which one.  You can show the sequence with menu: Favorites… Sequence and choosing which to show, or command, for example: sequence #0.1

(2) in the sequence window menu, choose "Structure… Associations," and in the resulting dialog change the association of every model (every ensemble member) from "none" to that same sequence, then click OK.

(3) in the sequence window menu, choose "Headers… RMSD."  A histogram of the per-residue RMSDs will be shown above the sequence.

So far I have assumed the structures are already superimposed on each other.  If they are not, the RMSDs will be high because they are calculated from the current positions.  If the structures are not superimposed, however, you can easily superimpose them.  One way is to choose from the sequence window menu  "Structure… Match".  You could try using all the positions of the sequence (no match options checked) or starting with all positions but turning on the option to iterate the fit, perhaps with cutoff 2 A.  You can try the matching several times to see what looks good to you.  Once you have gotten the structures matched (superimposed) the way you like, then you could go ahead with further steps.

(4) if you just want to show the per-residue RMSDs with color and/or "worm" fatness on one of the structures, you can just do that with Render by Attribute (in menu under Tools… Structure Analysis), in which you would choose the residue attribute named mavRMSD.

(5) if you want to write out the values, you can also do that with Render by Attribute, with "File… Save Attributes" in its menu, again choosing the residue attribute named mavRMSD.  You would just do that for one of the structues, because all of the structures would have the same values as each other.

Again, these RMSDs are based on the CA atoms only.   If you only had 2 structures in the ensemble, it would just be the CA-CA distance at each position.  There is no simple (nonprogramming) way to use all the backbone atoms.
I hope this helps,
Elaine C. Meng, Ph.D. 
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco

On Jan 14, 2014, at 7:10 AM, cantini at cerm.unifi.it wrote:

> Dear users
> Does anyone knows how to calculate through CHIMERA the backbone rmsd per residue of a NMR ensembles of protein structures ?
> thank a lot
> Francesca Cantini

More information about the Chimera-users mailing list