Hi,
'atomicfluct' is best used following an RMS-fit to an averaged
structure. Assuming you want local fluctuations for each chain, and
since you have several chains, you should probably rms-fit and
calculate fluctuations for each one separately (unless you want some
sort of global fluctuation for the entire system). The best procedure
in my opinion is to run your calculation in two phases, one in which
the coords are stripped and imaged, the second in which you perform
the actual calc. The easiest way to do this IMHO is with cpptraj since
the 'autoimage' command takes a lot of the guesswork out of imaging.
Note that in your case since you have ions in-between your chains that
the atom numbering of the system will be altered after the strip
command (i.e. the second chain will start at residue 211 instead of
213 since the two intervening ions will be stripped). The following
examples assume you are using cpptraj:
1) strip and image the trajectory, calculate the average structure:
parm protein.prmtop
trajin md1.mdcrd 1 500 1
-
-
-
trajin md50.mdcrd 1 500 1
strip :WAT,Cl- outprefix strip
average chain1.pdb :1-210 pdb
average chain2.pdb :211-420 pdb
average chain3.pdb :421-630 pdb
trajout strip.md.nc netcdf
After this you will have a stripped topology 'strip.protein.prmtop'
that corresponds to the stripped traj 'strip.md.nc' and 3 PDB files
containing the average structure of each chain.
2) rms fit to average structure, calculate atomic fluctuations:
parm strip.protein.prmtop
trajin strip.md.nc
parm chain1.pdb
reference chain1.pdb parm chain1.pdb [chain1]
parm chain2.pdb
reference chain2.pdb parm chain2.pdb [chain2]
parm chain3.pdb
reference chain3.pdb parm chain2.pdb [chain2]
rms ref [chain1] :1-210 :1-210
atomicfluct out chain1.fluct.dat :1-210.CA byatom
rms ref [chain2] :211-420 :1-210
atomicfluct out chain2.fluct.dat :211-420.CA byatom
rms ref [chain3] :421-630 :1-210
atomicfluct out chain3.fluct.dat :421-630.CA byatom
Note this is intended as a guideline only - since I don't have access
to your system I can only guess at the best way to go about
calculating things. Hopefully this is somewhat helpful. Good luck!
-Dan
On Fri, Nov 30, 2012 at 7:16 AM, Sangeetha B <sangeetha.bicpu.edu.in> wrote:
> Dear amber users,
>
> I am simulating a trimeric protein a with flexible C-terminal region for 50
> ns. Each chain is 210 aa in length. Residues 211 and 212 are two structural
> Cl- ions.
> I am having imaging issues while extracting the structures and performing
> any analysis. I have extracted pdb structures with different image center
> options.
> Without any imaging, the chains are separated in different directions and
> with different options, the location of these chains vary.
>
> I went through the amber mailing list and with the its help, I finally
> found the image commands shown below gave reasonably good structures when
> visualized.
> Whereas, when I use this for analyzing the RMS fluctuations very strange
> behavior is seen.
>
> The ptraj script that I use for generating RMS fluctuation is as follows:
>
> trajin md1.mdcrd 1 500 1
> -
> -
> -
> trajin md50.mdcrd 1 500 1
> strip :WAT,Cl-
> center :1-210 mass
> image :1-210 bymask :1-210
> center :1-423 mass
> image :1-423 bymask :1-423
> center :1-632 mass
> image :1-632 bymask :1-632
> image origin center
> atomicfluct out rmsf_md50ns.out .CA byatom
>
>
> When I plot the output from this script (rmsf_with_imaging.jpg) there is a
> sudden increase in fluctuation at the beginning of the second chain which
> starts at 213.
> But when I don't use any imaging commands (rmsf_no_imaging.jpg), the rms
> fluctuation does not show any such sudden increase.
> I have attached the images of these plots.
>
> Why is this this strange behavior observed?
> How will I make sure which set of image center commands works better for my
> system.
>
> Any suggestions would be very helpful.
>
>
>
>
>
>
>
> --
> Thank you,
>
> With regards,
> B. Sangeetha
> Ph.D Scholar,
> Centre for Bioinformatics,
> Pondicherry University,
> Pondicherry,
> India
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-9119 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Dec 05 2012 - 14:00:02 PST
