RE: Constraint failure in RATTLE algorithm for atom 6!

From: Radak, Brian K (
Date: Tue Apr 26 2016 - 16:26:30 CDT

How did you construct the PSF? Was it with psfgen and/or VMD? Can you post an input script with the exact commands that you used?

Brian Radak
Postdoctoral Appointee
Leadership Computing Facility
Argonne National Laboratory

9700 South Cass Avenue, Bldg. 240
Argonne, IL 60439-4854
(630) 252-8643
From: [] on behalf of faride badalkhani []
Sent: Tuesday, April 26, 2016 4:22 PM
To: namd-l;
Subject: Re: namd-l: Constraint failure in RATTLE algorithm for atom 6!


It is an internal alkyne carbon!


On Wed, Apr 27, 2016 at 1:47 AM, faride badalkhani <<>> wrote:
Thanks for your answer and time.


On Tue, Apr 26, 2016 at 7:44 PM, Arash Azari <<>> wrote:
salam Farideh,

I am not expert in this field, but I think this question has been answered in the mailing-list.
Based on my previous experience, it simply indicates that the system did not well minimized before running the simulation. You need much longer minimization.
As I remember from your previous emails, you used 0.5 fs time step which is not the step which normally people use in their MD simulations; you should use something around 2 fs. Please have a look at user manual.
Please repeat everything with this time step and minimize the system much longer; it depends on the size of your system and the number of atoms.
You should perform: minimization, (gradual heating depending on what you have done and need) equilibration, and main simulation. In some cases it is recommended to minimize the system a couple of times followed by equilibration.
I hope this helps.
One more thing; please do not post too much question to the mailing list as people will ignore very frequent questions.
Probably you can get much better help from your colleagues at your research group (Department).
Please let me know if you solve the problem.
Good luck!

Best regards,

On Mon, Apr 25, 2016 at 5:42 PM, faride badalkhani <<>> wrote:
Dear NAMD users,

I am trying to simulate a hyperbranched polymer in a water box. I defined the patches and built the Protein Structure File (PSF) using a tcl command, and there was nothing strange in psf file format. when I run a NVT simulation for 500 ps I got this error message:

ERROR: Constraint failure in RATTLE algorithm for atom 6!
ERROR: Constraint failure; simulation has become unstable.
ERROR: Exiting prematurely; see error messages above.

So, I reviewed NAMD mailing list and checked periodic cell size. I had minimized the system for 25 ps. I tried looking at my psf and pdb files in VMD. There was no atom with (0, 0, 0) coordinates.
I set the MARGIN parameter to 5 in the NAMD configuration file.

But I got the same error. Therefore, I tried the simulated annealing in vacuum. But, it does not work, too.

Atom 6 (in fact, 5) is an internal alkyne carbon! and the residue has a shape of

RESI POC -0.235 !
ATOM HA H 0.421
ATOM O1 OH1 -0.899 !
ATOM C1 CT2 0.198 ! H1
ATOM H1 HA2 0.090 ! __ |
ATOM H2 HA2 0.090 ! H3--C3==C2--C1--O1--HA
ATOM C2 CGT1 -0.015 ! |
ATOM C3 CGT2 -0.410 ! H2
ATOM H3 HGP 0.290 !
BOND H3 C3 C3 C2 C2 C1 C1 O1
BOND C1 H1 C1 H2 O1 HA

Any help will be appreciated.


This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:22:03 CST