On Mon, 2014-10-27 at 06:25 -0700, Caitlin Scott wrote:
> Hi Everyone,
>
> I tried using the ParmEd and the '-xpsf' option in chamber to generate my
> psf files.
>
> ParmEd: chamber -top top_all36_prot-mod.rtf -param par_all36_prot-mod.prm
> -psf step2_solvator.psf -crd step2_solvator.crd -nocmap -box
> 103.,103.,103.
This worked OK? (As in, it printed a topology file and incprd file?)
> Chamber: /usr/local/bin/chamber -top top_all36_prot-mod.rtf -param
> par_all36_prot-mod.prm -xpsf step2_solvator.psf -crd step2_solvator.crd
> -nocmap -vmd -box 103. 103. 103.
The -vmd option here does _not_ mean the PSF file came from VMD. It
means that you want chamber to print out a prmtop file that VMD will
read as an Amber topology file. To me this doesn't serve much purpose,
since you can just use the PSF file with VMD instead of the prmtop.
You certainly can't use the VMD-compatible prmtop you generated with
chamber here for actual simulations.
> Then I perform a minimization, but I have trouble with the equilibration
> step. I get the following error messages:
> vlimit exceeded for step 0; vmax = 1208.7464
> vlimit exceeded for step 0; vmax = 1208.7080
> vlimit exceeded for step 0; vmax = 1208.8334
> vlimit exceeded for step 0; vmax = 1208.9473
> vlimit exceeded for step 0; vmax = 1208.8851
Looks like you have overlapping particles. These velocities are _huge_,
indicating very large forces. You can use the "checkoverlap" command in
cpptraj to see if any particles are too close to each other.
One of the most common causes for this is that the unit cell dimensions
are not set correctly. In ParmEd, try using "-box bounding" instead of
"-box 103,103,103" to make sure that the box is defined to be big
enough. If you are using CHARMM GUI (which it appears you are?), I
believe overlapping particles is fairly common. In this case, you may
need to make your box a little bigger to let particles move out of the
way, and your minimization protocol probably has to be done very
carefully (and carefully check the resulting minimized structure before
using it to heat or equilibrate).
Another possibility is that you asked for a non-orthorhombic box, like a
truncated octahedron or a rhombic dodecahedron. In that case, you need
to specify the angles between the unit cell vectors as well as the cell
lengths.
>
> =====================================================================================
> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
> = EXIT CODE: 11
> = CLEANING UP REMAINING PROCESSES
> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
> =====================================================================================
> APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)
>
> I don't have this problem when I generate the psf file using the '-psf'
> option.
> Chamber: /usr/local/bin/chamber -top top_all36_prot-mod.rtf -param
> par_all36_prot-mod.prm -psf step2_solvator.psf -crd step2_solvator.crd
> -nocmap -vmd -box 103. 103. 103.
If you are using chamber, either the -xpsf or the -psf flag should work:
not both. The difference between these two is that XPLOR PSF files
assign atom types using the type label, whereas the CHARMM PSF file uses
the type index.
ParmEd auto-detects the format of the PSF... chamber does not.
One last question -- is there a reason you're disabling CMAPs? Unless
you are specifically trying to test the effect that removing CMAPs has
on the simulation, this is typically discouraged.
Hope this helps,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 27 2014 - 07:00:02 PDT
