Hi George, My non-mass-weighted script should have been dividing by the number of atoms and it wasn’t. I’ve attached a corrected version. Nonetheless, the Amber calculation is using mass weighting, so that is not the issue. I need to emphasize again that the script is going to compute the radius of gyration of all atoms in your model, regardless of which ones are selected. The only way to compute the value (with that script) for a subset of atoms is to delete all the other ones. For instance, with the PDB you sent if I delete all atoms except for residue 26, this is the output I get using the mass-weighted script: #0 center of mass: 26.1014 36.0445 34.9565, radius of gyration: 1.934 If you want the script to work on the currently selected atoms instead of all atoms, I could send you a modified script that does that. Let me know if you want that. —Eric
On Apr 10, 2017, at 12:12 PM, George Tzotzos <gtzotzos@me.com> wrote:
Hi Eric,
Thank you very much. I tried the new script on residues 126 and 127 of the attached PDB file. Both ligands are indole.
The Reply Log returns
#0 center of mass: 29.5741 29.9514 29.7281, radius of gyration: 3010.778 #0 center of mass: 29.5741 29.9514 29.7281, radius of gyration: 3010.778
This looks unrealistic and seems to be off by at least 3 orders of magnitude.
I used the following script based on AmberTools15
parm cmplx_solv.prmtop trajin prod_0-100ns.nc radgyr :126&!(@H=) out RoG.dat mass nomax run
I get a radius of gyrations with mean ~ 1.8 angstrom. See snapshot attached.
For easy reference, the AmberTools15 manual states (p.586)
radgyr | rog
radgyr [name>] [<mask>] [out <filename>] [mass] [nomax] [tensor]
[<name>] Data set name. [<mask>] Atoms to calculate radius of gyration for; default all atoms. [out <filename>] Write data to <filename>.
[mass] Mass-weight radius of gyration. [nomax] Do not calculate maximum radius of gyration. [tensor] Calculate radius of gyration tensor, output format ’XX YY ZZ XY XZ YZ’. Data Sets Created:
Data Sets Created:
<name> Radius of gyration in Ang. <name>[Max] Max radius of gyration in Ang. <name>[Tensor] Radius of gyration tensor; format ’XX YY ZZ XY XZ YZ’.
Calculate the radius of gyration of specified atoms. For example, to calculate only the mass-weighted radius of gyration (not the maximum) of the non-hydrogen atoms of residues 4 to 10 and print the results to “RoG.dat”:
radgyr :4-10&!(@H=) out RoG.dat mass nomax
<PastedGraphic-1.png>
<indX2.pdb>
On 10 Apr 2017, at 20:30, Eric Pettersen <pett@cgl.ucsf.edu <mailto:pett@cgl.ucsf.edu>> wrote:
Hi George, Mathematically, radius of gyration is pretty simple. From Wikipedia: radius of gyration is the root mean square distance of the object's parts from its center of mass. Looking at the script, perhaps the radius of gyration computation should not be mass weighted. The script I’ve attached eliminates that mass weighting. Let me know if that gives the results you expect.
—Eric
Eric Pettersen UCSF Computer Graphics Lab
<radius-of-gyration.py>
On Apr 7, 2017, at 11:16 AM, George Tzotzos <gtzotzos@me.com <mailto:gtzotzos@me.com>> wrote:
Hi Elaine,
The reason I asked is because I’ve measured the radius of gyration of a particular ligand during the course of an MD trajectory using AmberTools. Comparing the mean radius of gyration of the ligand during the trajectory with the corresponding value generated by the Chimera script, the difference is by a factor of 4. AmberTools radius of gyration is also in angstrom.
Have a good day
George
On 7 Apr 2017, at 19:58, Elaine Meng <meng@cgl.ucsf.edu <mailto:meng@cgl.ucsf.edu>> wrote:
Hi George, Units would be the same as your atomic structure coordinates, so yes, angstroms if you are just using “regular” PDB files. I’m not sure why you would expect different units. Best, Elaine
On Apr 7, 2017, at 10:48 AM, George Tzotzos <gtzotzos@me.com <mailto:gtzotzos@me.com>> wrote:
Thank you Eric,
The problem was exactly what you identified. I assume the units are in Angstrom
George
On 7 Apr 2017, at 01:46, Eric Pettersen <pett@cgl.ucsf.edu <mailto:pett@cgl.ucsf.edu>> wrote:
Hi George, First, that script measures the radius of gyration for all the atoms in a model, not just ligands atoms or some subset of the model’s atoms. So, my best guess is that the model that’s from your MD simulation has a lot solvent atoms that Chimera isn’t displaying by default. So you should check that there are the same number of atoms in both models. One way to do that is with the command:
sel #0 (number of selected atoms [all of model 0] shown in the status line) sel #1 (same thing for model 1).
If the difference in the models is the solvent you would want to use the command “delete solvent” before running the gyration script.
—Eric
Eric Pettersen UCSF Computer Graphics Lab
> On Apr 6, 2017, at 12:35 PM, George Tzotzos <gtzotzos@me.com <mailto:gtzotzos@me.com>> wrote: > > Taking a reference PDBID 3n7h, I’ve generated for an MD simulation topology and restart files. > > I’ve loaded on Chimera 3n7h and then the starting frame of the MD simulation. > Used the MatchMaker and found perfect superposition of the protein backbone and two ligands (3n7h is #0 and the ligands #0:129.A and #0.132.B > > I run the radius of gyration script (see attached). I found that the radii of gyration for the ligands in the two models are completely different. > > #0 center of mass: 12.6543 10.4356 15.4086, radius of gyration: 21.182 > #1 center of mass: 42.7381 42.5171 42.6163, radius of gyration: 37.390 > > I don’t follow why this is so. Any explanation would be most welcome. > > A second question regards the units of the radii of gyration. Are they Angstrom or nm? > > Thank you in advance > > George > > <rad_gyration.py>_______________________________________________ > Chimera-users mailing list: Chimera-users@cgl.ucsf.edu <mailto:Chimera-users@cgl.ucsf.edu> > Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users <http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu <mailto:Chimera-users@cgl.ucsf.edu> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users <http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>
_______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu <mailto:Chimera-users@cgl.ucsf.edu> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users <http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>