[gmx-users] Re: grompp for minimization: note & warning

[Justin Lemkul]

On 9/17/13 7:02 AM, shahab shariati wrote:
> Dear Justin
>
> Thanks for your reply
>
>
>> You're still working with outdated software.  Please follow my
>> suggestion of trying 4.6.3.
>
> I try to install 4.6.3.
>
>> The larger issue, in the end, is the unphysical configuration of the drug
>> in the membrane.  Independent of Gromacs version, that setup will
>> always fail.
>
> I did minimization in 2 other states:
>
> 1) without drug (only membrane: DOPC nad cholesterol).
>
> 2) with drug but, this time, I put drug into water molecules not into
> membrane.
>
> Then I did mdrun for both of states. In both of state, there are not
> following issues:
>
> Warning: 1-4 interaction between 434 and 407 at distance 3.023 which is
> larger than the 1-4 table size 2.200 nm These are ignored for the rest of
> the simulation This usually means your system is exploding,
>
> if not, you should increase table-extension in your mdp file
> or with user tables increase the table size
>
> step 23: Water molecule starting at atom 10613 can not be settled.
> Check for bad contacts and/or reduce the timestep if appropriate.
> Wrote pdb files with previous and current coordinates
>
> --------------------------------------------------------------------------
> There are following outputs:
>
> for state 1)
> -------------
>
> Reading file em.tpr, VERSION 4.5.1 (single precision)
> Starting 4 threads
> Making 2D domain decomposition 1 x 2 x 2
>
> Steepest Descents:
>     Tolerance (Fmax)   =  1.00000e+03
>     Number of steps    =        50000
>
> Stepsize too small, or no change in energy.
> Converged to machine precision,
> but not to the requested precision Fmax < 1000
>
> Double precision normally gives you higher accuracy.
> You might need to increase your constraint accuracy, or turn
> off constraints alltogether (set constraints = none in mdp file)
>
> writing lowest energy coordinates.
>
> Steepest Descents converged to machine precision in 881 steps,
> but did not reach the requested Fmax < 1000.
> Potential Energy  =  2.1828770e+05
> Maximum force     =  1.3656898e+04 on atom 618
> Norm of force     =  5.1748779e+02
> --------------------------------------------------------------------
>

As I suggested before, visualize the output file and figure out what is around 
atom 618 to cause such forces.

> for state 2)
> ----------------
>
> Reading file em.tpr, VERSION 4.5.1 (single precision)
> Starting 4 threads
> Making 2D domain decomposition 1 x 2 x 2
>
> Back Off! I just backed up em.trr to ./#em.trr.1#
>
> Back Off! I just backed up em.edr to ./#em.edr.1#
>
> Steepest Descents:
>     Tolerance (Fmax)   =  1.00000e+03
>     Number of steps    =        50000
>
> Stepsize too small, or no change in energy.
> Converged to machine precision,
> but not to the requested precision Fmax < 1000
>
> Double precision normally gives you higher accuracy.
> You might need to increase your constraint accuracy, or turn
> off constraints alltogether (set constraints = none in mdp file)
>
> writing lowest energy coordinates.
>
> Back Off! I just backed up em.gro to ./#em.gro.1#
>
> Steepest Descents converged to machine precision in 845 steps,
> but did not reach the requested Fmax < 1000.
> Potential Energy  =  1.9664772e+05
> Maximum force     =  1.0986521e+04 on atom 241
> Norm of force     =  4.4174323e+02
> ---------------------------------------------------------------------------
>

Same advice as above.

> When I see created gro file from minimization (em.gro), I see some dopc and
> cholesterol molecules were broken and they devided 2 or 3 parts. Input gro
> file (system.gro) has 1835 residues. Created gro file from minimization
> (em.gro) has 1835 residues. But when I load em.gro file in VMD, there are
> 1950 residues in graphical representation --> selections --> residues part.
>

What VMD thinks is a fragment, molecule, or residue is basically irrelevant. 
It's guessing based on guessing connectivity.  Your topology is definitive; VMD 
is not.

> Can I deduce my problem has not dependent to presence of drug?
>

There seem to be problems in both cases.  Solve the issue with the membrane first.

> Are there problem about my force field parameters or topology file ?
>
> I get force field parameters from lipid book (for dopc and cholesterol) and
> from prodrg server (for drug).
>

PRODRG generates notoriously bad parameters, particularly the charges.  See 
http://pubs.acs.org/doi/abs/10.1021/ci100335w.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441

==================================================