Re: Energy increase with fixed atoms in POPC membrane

From: Tristan Croll (tristan.croll_at_qut.edu.au)
Date: Fri Feb 19 2016 - 10:51:52 CST

I find AutoIMD can be a godsend in these situations. Once you learn how to use it, you can run a quick minimisation of just the area around your problem atoms, while watching exactly what's going on.

________________________________
From: owner-namd-l_at_ks.uiuc.edu <owner-namd-l_at_ks.uiuc.edu> on behalf of Norman Geist <norman.geist_at_uni-greifswald.de>
Sent: Friday, 19 February 2016 5:47 PM
To: 'Mihaela Drenscko'
Cc: namd-l_at_ks.uiuc.edu
Subject: AW: namd-l: Energy increase with fixed atoms in POPC membrane

Try to increase minimization steps. But sometimes even the minimizer can’t get rid of all bad contacts, sometimes shown in the out similar to “GIVING UP ON XXX ATOMS WITH BAD CONTACTS”. In this case you should resolve the superimposed atoms by hand using VMD and the mouse->move tool. Also make sure the periodic box size is set correctly.

Generally you should leave the minimization with an total energy less then 9999999999.9999, in the best case it did converge to a constant.

Norman Geist

Von: Mihaela Drenscko [mailto:quo.physics_at_gmail.com]
Gesendet: Freitag, 19. Februar 2016 02:08
An: Norman Geist <norman.geist_at_uni-greifswald.de>
Betreff: Re: namd-l: Energy increase with fixed atoms in POPC membrane

Hi,

I used Dynamic Bonds representation with 0.9 cutoff, found the close atoms, unfixed them (tcl script enclosed), and still I've got now an error about the RATTLE algorithm (or maybe the script is not OK?):
 ENERGY: 1000 4020.3417 4438.9290 3471.1582 11.6114 -20100.9995 9999999999.9999 0.0000 0.0000 0.0000 9999999999.9999 0.0000 9999999999.9999 9999999999.9999 0.0000 9999999999.9999 9999999999.9999 160380.0000 9999999999.9999 9999999999.9999

WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 1000
OPENING COORDINATE DCD FILE
WRITING COORDINATES TO DCD FILE AT STEP 1000
WRITING COORDINATES TO RESTART FILE AT STEP 1000
FINISHED WRITING RESTART COORDINATES
The last position output (seq=1000) takes 0.001 seconds, 109.023 MB of memory in use
WRITING VELOCITIES TO RESTART FILE AT STEP 1000
FINISHED WRITING RESTART VELOCITIES
The last velocity output (seq=1000) takes 0.000 seconds, 109.023 MB of memory in use
REINITIALIZING VELOCITIES AT STEP 1000 TO 300 KELVIN.
TCL: Running for 2500 steps
ERROR: Constraint failure in RATTLE algorithm for atom 11155!
ERROR: Constraint failure; simulation has become unstable.
ERROR: Exiting prematurely; see error messages above.

Is there anything else I could do?

Thank you,

Mihaela

On Thu, Feb 18, 2016 at 3:03 AM, Norman Geist <norman.geist_at_uni-greifswald.de<mailto:norman.geist_at_uni-greifswald.de>> wrote:

Hey,

energies being 9999999999.9999 means they are actually infinite. This is usually caused by superimposed atoms. As you fxed some parts of your system, the superimposed atoms are probably fixed and so can’t be minimized. Try to find the atoms which are too close, maybe using the “Dynamic Bonds” representation with a small cutoff like e.g. 0.9 and unfix them or remove the clash by hand using “VMD Main->Mouse->Move->Atom” for instance.

Norman Geist

Von: owner-namd-l_at_ks.uiuc.edu<mailto:owner-namd-l_at_ks.uiuc.edu> [mailto:owner-namd-l_at_ks.uiuc.edu<mailto:owner-namd-l_at_ks.uiuc.edu>] Im Auftrag von Mihaela Drenscko
Gesendet: Mittwoch, 17. Februar 2016 23:27
An: namd-l_at_ks.uiuc.edu<mailto:namd-l_at_ks.uiuc.edu>
Betreff: namd-l: Energy increase with fixed atoms in POPC membrane

Hi,

I was following the tutorial:

http://www.ks.uiuc.edu/Training/Tutorials/science/membrane/mem-tutorial.pdf

to build a POPC membrane 50x50 with PCL (instead of kcsa), using CHARMM36.

In section 3.1, page 38, it shows how fix atoms:

set fixed [atomselect top "water or name CLA POT or protein or

\

(chain L and name O2 P1 O3 O4 O1 C15 H52 H51 H11 C11 H12

\

N C14 H42 H43 H41 C12 H22 H23 H21 C13 H33 H31 H32)"]

For the membrane pdb in charmm36, there are other notaions for atoms. I made this change:

set fixed [atomselect top "water or name CLA POT or protein or \
(chain L and name O11 P O13 O14 O12 C11 H11B H11A H12A C12 H12B \
N C15 H15B H15C H15A C13 H13B H13C H13A C14 H14C H14A H14B)"]

Was this replacement wrong? When I run the jib, the energy does not minimize (I increased minimization steps to 5000, decreased timestep 1 fs, and run it for 2500 steps):

ENERGY: 7500 4446.6430 6773.7383 3764.0810 41.4892 -20222.3357 9999999999.9999 0.0000 0.0000 4719.5622 9999999999.9999 297.2827 9999999999.9999 9999999999.9999 295.0647 9999999999.9999 9999999999.9999 160380.0000 9999999999.9999 9999999999.9999

WRITING EXTENDED SYSTEM TO OUTPUT FILE AT STEP 7500
CLOSING EXTENDED SYSTEM TRAJECTORY FILE
WRITING COORDINATES TO OUTPUT FILE AT STEP 7500
CLOSING COORDINATE DCD FILE
The last position output (seq=-2) takes 0.001 seconds, 111.801 MB of memory in use
WRITING VELOCITIES TO OUTPUT FILE AT STEP 7500

Thank you,

Mihaela

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:11 CST