[gmx-users] Group-specific differences in RMS fluctuation results via g_rmsf

[Keith Ball]

I would like to calculate the RMS fluctuations of residues in an active
site, using the following command and options:

 

g_rmsf -f x16traj.trj -s x16topol.tpr -n index2.ndx -b 1000 -res -o
rmsf.xvg -od rmsdev.xvg

 

When the program asks me to "Select group(s) for root mean square
calculation", 

I select the appropriate group number. This gives me rmsf.xvg and
rmsdev.xvg files which only contain values for the residues specified in
the group I selected. I am focusing on rmsf.xvg, as that has the RMS
fluctuation data.

 

I have groups defined for the entire active site, as well as several
sub-regions of it. Several residues occur in more than one sub-region.

 

I notice that the values I get for RMSF for the same residue differ,
depending on which group I have chosen. (i.e. value for a given residue,
when the whole pocket is selected as the group, is different from when I
only select a sub-pocket containing that residue).

 

Is g_rmsf somehow referring the coordinates of this residue's atoms to a
constantly-changing reference frame? It seems to me that the residue's
RMSF(as an average of the RMSF of its atoms) should only be in reference
to the average position of its atoms, and not dependent upon some
external "frame".

 

Should I just use the whole protein as the reference, and extract the
residue values of interest by hand?

 

Thanks,

  Keith

============================
Keith D. Ball, PhD
Staff Scientist
Catalyst Biosciences
290 Utah Ave
S. San Francisco, CA 94080

650-266-8661 (phone)
650-871-2475 (fax)
kball at catbio.com

http://www.catbio.com
============================

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20061203/06fc46f6/attachment.html>