From: Peter Freddolino (
Date: Sat Jul 28 2007 - 07:54:35 CDT

Hi Karol,
VMD needs at least 4-5 atoms for a reliable fit, and preferrably more.
However, you're performing your fit with the heavy atoms in a single
side chain, which is probably unreliable unless it's a large one. I'd
recommend one of the following two options:
-Do the fitting based on the entire protein's backbone, and only the
rmsd calculation using the side chain only. This would be more in
agreement with usual rmsd conventions.


-Include the heavy atoms of the residue's backbone in the fit, to give
it enough to work with

When using commands that move things based on a fit, it is also usually
best to move the entire molecule, even if you only care about one
specific part of it.


Karol Kaszuba wrote:
> Hello,
> I would like to report some strange problem.
> I performed MD.
> I loaded the PSF file.
> I loaded the DCD file.
> I calculated RMSD for heavy atoms of one SIDECHAIN and I noticed the
> following problem:
> The RMSD value was irrationally high - reached even 350 A.
> The carbon atoms of my sidechain were 300 A away from the protein.
> The hydrogen atoms of my sidechain were "in place" (still constituted a
> part of the protein).
> First I thought that my simulation went abnormally - for example: heavy
> atoms of the sidechain "escaped" and remained separated but they didn't -
> I was observing them during DCD playing.
> It looks like RMSD calculation moves my heavy atoms away from protein.
> Furthermore - when I calculated RMSD for whole resid 37 I obtained a
> correct rmsd value and the heavy atoms of the sidechain were not moved for
> 300 A.
> I am using the folowing rmsd.tcl script with the VMD for linux 32, version
> 1.8.5
> set outfile [open rmsd_37.dat w]
> set nf [molinfo top get numframes]
> set frame0 [atomselect top "protein and resid 37 and sidechain and not
> hydrogen" frame 0]
> set sel [atomselect top "protein and resid 37 and sidechain and not
> hydrogen"]
> # rmsd calculation loop
> for {set i 1 } {$i < $nf } { incr i } {
> $sel frame $i
> $sel move [measure fit $sel $frame0]
> puts $outfile "[measure rmsd $sel $frame0]"
> }
> close $outfile
> I would like to ask if someone know what may cause this problem and how to
> omit it?
> Thank you in advance,
> Regards,
> Karol

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:00 CST