Re: [modeller_usage] Averaged RMSD for 100 models

[Modeller Caretaker <>]

Robert Selwyne wrote:
> I had run MODELLER to produced model 1-100.pdb (100 models) of same acid 
> sequence but different energy values. Now, i want to calculate the RMSD 
> values for the models. I know that SUPERPOSE will do for me. But i 
> couldnot correlate and implement the results.

You can align multiple structures with alignment.compare_structures() 
(see http://salilab.org/modeller/manual/node193.html), 
alignment.malign3d() (see 
http://salilab.org/modeller/manual/node195.html) or alignment.salign().

> Say in VMD 1.8.4 in the RMSD calculator option, if *Average* is 
> selected, VMD first computes the average x, y, z coordinates of the 
> selected atoms in each molecule, then computes the RMS distance of each 
> molecule from that average structure.

This is essentially the same algorithm used by alignment.malign3d(). See 
the manual linked above.

There is an example in examples/commands/malign3d.py. If you are using 
automodel to build your models, you can ask it to automatically align 
the generated models and calculate many properties, including the RMS, 
by setting final_malign3d to True. See examples/automodel/model-fast.py 
for an example.

	Ben Webb, Modeller Caretaker
--
             http://www.salilab.org/modeller/
Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage