RE: Temperature reassignment

From: Tristan Croll (tristan.croll_at_qut.edu.au)
Date: Sun Jan 12 2014 - 21:32:03 CST

I'm not convinced the problem is in temperature. A few other possible problems you might want to consider:

- Are you using any unusual topologies/patches - and if so, are you missing a bond or angle specification somewhere? In such cases the offending atoms will fluctuate wildly and quickly crash your simulation.

- Have you included enough padding in the Z direction? With membrane simulations you typically need a lot more water padding in the Z direction to make up for the lack of flexibility in the X-Y plane as the system does its initial contraction. If you don't have enough the Z axis will contract to the point where your protein crashes into its periodic twin.

A few other points/suggestions:

- 500,000 minimization steps seems a bit excessive. I rarely use more than 5,000 steps myself

- Try running the simulation from where the minimization left off, with restarts every 50-100 steps. This will give you a chance to visualise what's happening to the offending atoms.

Good luck!

Tristan

From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf Of Thomas Evangelidis
Sent: Friday, 3 January 2014 1:54 AM
To: PANAGIOTA KYRIAKOU
Cc: namd-l
Subject: Re: namd-l: Temperature reassignment

Another possible error might be that your membrane is not equilibrated -it is not obvious in your script whether you did it before the step that fails. Typically, in protein-membrane systems we follow several stages of equilibration. The following is an outline of the stages I follow with the lipid11 force field that requires constant surface tension (NPgT ensemble):

# Stage 1: minimization of lipid tails only, the rest are kept fixed.
# Stage 2: minimization of the whole system apart from the protein which is constrained to its initial configuration. Then the system is slowly heated up to 320 K and
# a small equilibration at the NPgT ensemble of 0.5 ns is conducted with the protein constrained.
# Stage 3: NPgT equilibration with no constraints.
In your case (CHARMM ff) the surface tension is not necessary. Also the final temperature depends of the lipid constitution of the membrane. Have a look at the NAMD protein-membrane tutorial for more info...

On 2 January 2014 02:17, PANAGIOTA KYRIAKOU <kyria008_at_gmail.com<mailto:kyria008_at_gmail.com>> wrote:
Hi there,
Thanks for your answer..
I thought the use of the commands reassignFreq, reassignTemp etc was gradual heating. I just do not understand what I am doing wrong.
I had problems with the constrains so this is why I used 1fs.

Giota

Panagiota Kyriakou

Graduate Student

Dept. of Chemical Engineering and Materials Science

University of Minnesota

email: kyria008_at_umn.edu<mailto:kyria008_at_umn.edu>

On Wed, Jan 1, 2014 at 9:27 AM, Thomas Evangelidis <tevang3_at_gmail.com<mailto:tevang3_at_gmail.com>> wrote:
Energy minimization is like dropping the temperature of the system to 0 Kelvin. Therefore, in order to initialize velocities at 303 Kelvin you must first do the gradual heating from 0->303 K otherwise the system will blow up. Unlike GROMACS or AMBER, NAMD does not do the heating automatically. Replace "run 1000000" with something like:

    # Heating
    set HeatSteps 5
    set HeatMDsteps 1000
    for {set i 1} {$i <= $HeatSteps } { incr i } {
        set t [expr $temp*$i/$HeatSteps]; puts "Heat structure to $t K for $HeatMDsteps steps."
        reinitvels $t
        run $HeatMDsteps
    }
    reinitvels $temp
    run 1000000 ;# 1 ns at $temp Kelvin

Also are you sure you want to use a timestep of 1 fs?

On 31 December 2013 19:28, PANAGIOTA KYRIAKOU <kyria008_at_gmail.com<mailto:kyria008_at_gmail.com>> wrote:
Hello,

I have been trying to equilibrate my system that consists of protein embedded in a membrane and I get the following error:

TIMING: 500000 CPU: 9750.65, 0.021551/step Wall: 9750.65, 0.021551/step, 0 hours remaining, 556.101562 MB of memory in use.
ETITLE: TS BOND ANGLE DIHED IMPRP ELECT VDW BOUNDARY MISC KINETIC TOTAL TEMP POTENTIAL TOTAL3 TEMPAVG PRESSURE GPRESSURE VOLUME PRESSAVG GPRESSAVG
ENERGY: 500000 48376.8664 114689<tel:8664%20%C2%A0%20%C2%A0114689>.7899 11643.7243 506.4399 -580965.4098 103191<tel:4098%20%C2%A0%20%C2%A0103191>.7767 0.0000 246.2954 0.0000 -302310.5171 0.0000 -302310.5171 -302310.5171 0.0000 6488.4885 6928.4046 1914140.8650 6488.4885 6928.4046

WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 500000
WRITING COORDINATES TO DCD FILE AT STEP 500000
WRITING COORDINATES TO RESTART FILE AT STEP 500000
FINISHED WRITING RESTART COORDINATES
WRITING VELOCITIES TO RESTART FILE AT STEP 500000
FINISHED WRITING RESTART VELOCITIES
TCL: Writing to files with basename A_equil_1_min.
WRITING EXTENDED SYSTEM TO OUTPUT FILE AT STEP 500000
WRITING COORDINATES TO OUTPUT FILE AT STEP 500000
WRITING VELOCITIES TO OUTPUT FILE AT STEP 500000
TCL: Running for 1000000 steps
REASSIGNING VELOCITIES AT STEP 500000 TO 303.15 KELVIN.
colvars: Writing the state file "A_equil_1.restart.colvars.state".
colvars: Synchronizing (emptying the buffer of) trajectory file "A_equil_1.colvars.traj".
ERROR: Constraint failure in RATTLE algorithm for atom 6522!
ERROR: Constraint failure; simulation has become unstable.
ERROR: Constraint failure in RATTLE algorithm for atom 13834!
ERROR: Constraint failure; simulation has become unstable.
ERROR: Constraint failure in RATTLE algorithm for atom 1812!
ERROR: Constraint failure; simulation has become unstable.

When I tried to fix it I realized that the Temperature increased from 0 to 312 K, instantly after minimization even though I am using temperature reassignment. Here is my configuration file:

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
# Custom variables
set temp 303.15

# inputs
  structure ../A.psf ;
  coordinates ../A.pdb ;
paraTypeCharmm on
parameters ../../common/topology/par_all36_prot_nomass.prm
parameters ../../common/topology/par_all36_water_ions.prm
parameters ../../common/topology/par_all36_lipid.prm

temperature $temp ;
# simulation parameters
outputname A_equil_1 ;
outputEnergies 1000 ;
outputTiming 1000 ;
xstFreq 10000 ;
dcdFreq 10000 ;
dcdUnitCell yes
binaryoutput on ;
restartfreq 10000 ;
seed 1369948321 ;

# calculation parameters
exclude scaled1-4 ;
1-4scaling 1.0 ;
switching on ;
switchDist 9 ;
cutoff 10 ;
pairlistdist 12 ;
stepsPerCycle 20 ;
rigidbonds all ;
vdwForceSwitching yes ;
timestep 1 ;
nonBondedFreq 1 ;
fullElectFrequency 1 ;
cellBasisVector1 139.65999603271484 00.00 00.00 ;
cellBasisVector2 00.00 140.35699462890625 00.00 ;
cellBasisVector3 00.00 00.00 97.64900207519531 ;
cellorigin -0.9953396916389465 4.457383155822754 2.071026086807251 ;
wrapAll on ;
wrapWater on ;
wrapNearest off ;

# PME
Pme on ;
PmeGridsizeX 144 ;
PmeGridsizeY 150 ;
PmeGridsizeZ 108 ;

# Control (need to be specified before runs)
useFlexibleCell yes ;
useGroupPressure yes ;
useConstantRatio yes ;

# constant pressure
langevinPiston on
langevinPistonTarget 1.01325
langevinPistonPeriod 50.0
langevinPistonDecay 25.0
langevinPistonTemp $temp

reassignFreq 1000 ;
reassignTemp 1 ;
reassignIncr 1 ;
reassignHold $temp ;

colvars on
colvarsConfig ../restraints/A_equil_1.col

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
# EXECUTION
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~#
minimize 500000

run 1000000

Do you know what could have caused the increase in the temperature?
I would appreciate your help so much,

Giota

Panagiota Kyriakou

Graduate Student

Dept. of Chemical Engineering and Materials Science

University of Minnesota

email: kyria008_at_umn.edu<mailto:kyria008_at_umn.edu>

--
======================================================================
Thomas Evangelidis
PhD student
University of Athens
Faculty of Pharmacy
Department of Pharmaceutical Chemistry
Panepistimioupoli-Zografou
157 71 Athens
GREECE
email: tevang_at_pharm.uoa.gr<mailto:tevang_at_pharm.uoa.gr>
          tevang3_at_gmail.com<mailto:tevang3_at_gmail.com>
website: https://sites.google.com/site/thomasevangelidishomepage/
--
======================================================================
Thomas Evangelidis
PhD student
University of Athens
Faculty of Pharmacy
Department of Pharmaceutical Chemistry
Panepistimioupoli-Zografou
157 71 Athens
GREECE
email: tevang_at_pharm.uoa.gr<mailto:tevang_at_pharm.uoa.gr>
          tevang3_at_gmail.com<mailto:tevang3_at_gmail.com>
website: https://sites.google.com/site/thomasevangelidishomepage/

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:02 CST