[gmx-users] solvation_box_preparation

Wed Feb 9 13:53:28 CET 2011

shahid nayeem wrote:
> I ran normal MD and it runs fine. Today I used six 10ps MD keeping
> temperature 300 K and each time gradually increasing pressure from 1
> bar to 5 bar to 25, 50, 75 and then 100bar. It all ran successfully.
> Thereafter doing simulated annealing for cooling and heating and
> equilibration. It all ran successfully without any LINC warning. But
> should I trust this box because you said parameters assigned by prodrg
> are none-sense.

I've made my thoughts on that matter pretty clear, I think.  You have to justify 
to reviewers that your parameters are accurate.  We recently published a paper 
detailing the artifacts that can arise from using unrefined PRODRG topologies; 
the incorrect results are quite pronounced.  You need to derive suitable charges 
in conjunction with sensible charge groups.  Garbage in, garbage out.

-Justin

> shahid Nayeem
> 
> On Wed, Feb 9, 2011 at 5:26 PM, Justin A. Lemkul <jalemkul at vt.edu> wrote:
>>
>> shahid nayeem wrote:
>>> Hi Justin
>>> Thanks a lot.
>>> I tried doing energy minimization and th lowering emtotal to 200 and
>>> the system converged to
>>> Steepest Descents converged to Fmax < 200 in 1411 steps
>>> Potential Energy  = -3.9063050e+05
>>> Maximum force     =  1.3442458e+02 on atom 927
>>> Norm of force     =  1.4758101e+01
>> OK, so energy minimization was fine.
>>
>>>  For equilibration of solvation box I am following a biophys j paper
>>> in which protocol for urea box preparation is given. A simulated
>>> annealing under high pressure (ref_p=100) to cool system from 300 to
>>> 0K thereafter heating back to 300K at ref_p=1. Again 1ns MD at same
>>> condition. If this way is harsh treatment of system then suggest me
>>> the way out.
>>> My sa.mdp sa_hot.mdp and sa_equilibriation.mdp as well as chaps.itp
>>> are attached with this mail
>> There's nothing obviously wrong with the .mdp file.  Have you tried running
>> "normal" MD to check for stability?  I suggested that before, and I'm still
>> curious.
>>
>> Also, your CHAPS topology makes no sense.  The haphazard charge assignment
>> and nonsensical charge groups indicate to me that PRODRG has done its usual
>> job of assigning funny parameters.  Even if you could get the simulations to
>> run, they shouldn't be trusted, and bad parameters could well be the source
>> of your problem.  Parameterization is time-consuming and difficult, but so
>> is generated quality, usable data :)
>>
>> -Justin
>>
>>> Shahid Nayeem
>>> On Mon, Jan 31, 2011 at 9:47 AM, Justin A. Lemkul <jalemkul at vt.edu> wrote:
>>>> shahid nayeem wrote:
>>>>> Please tell me where I am wrong. I downloaded pdb of chaps and used
>>>>> prodrg server to get .itp and .gro file. Then I checked .itp for any
>>>>> missing charge and I found it correct. Then I created 6.0x6.0x6.0 box
>>>> PRODRG doesn't have a problem of missing charges.  It provides
>>>> notoriously
>>>> incorrect charges.
>>>>
>>>>> with genbox inserting 7 molecules of chaps.gro. Then again using
>>>>> genbox and -maxsol I put 510 spc.itp in the box to get a density
>>>>> approaching 1. Then I did steepest descent energy minimization with
>>>>> constraints = none, for emtotal=2000 and emstep=3000. Up to this the
>>>> These settings make no sense.  An emtol of 2000 is very high, and emstep
>>>> of
>>>> 3000 is total nonsense.  How well did you EM converge?  What were the
>>>> values
>>>> of the potential energy and maximum force?
>>>>
>>>>> gromacs runs fine. when I start simulated annealing for cooling at
>>>>> high pressure with constraint = all_bonds the programme gives fatal
>>>>> error linc warning and stops. If I do energy minimization with
>>>>> constraint =all_bonds then also with some error of linc wrning the
>>>>> minimization is completed. When I do minimization without adding water
>>>>> then there is no linc warning and minimization is completed but with
>>>>> final positive potential energy. Then as suggested by Justin I used
>>>>> smaller box and there also in simulated annealing stage the system
>>>>> gives linc warning and the programme stops with fatal error. Please
>>>>> tell me where I am wrong.
>>>> How about simplifying the problem.  Does the system run under normal
>>>> conditions?  In other words, can you run normal MD?  You're treating the
>>>> system very harshly with the combination of high pressure and annealing.
>>>>  Without seeing your .mdp file for this process, it's impossible to say
>>>> how
>>>> reasonable your settings are.
>>>>
>>>> It is also possible that your parameters for CHAPS (if they are the
>>>> default
>>>> ones from PRODRG) are incorrect.  The charges and charge groups nearly
>>>> always are. Without seeing them, there's nothing better to offer.
>>>>
>>>> -Justin
>>>>
>>>>> shahid nayeem
>>>>>
>>>>> On Fri, Jan 28, 2011 at 10:59 AM, Mark Abraham <Mark.Abraham at anu.edu.au>
>>>>> wrote:
>>>>>> On 28/01/2011 3:51 PM, shahid nayeem wrote:
>>>>>>> Thanks Justin
>>>>>>> I tried with new box size of 2.8x2.8x2.8 . During energy minimization
>>>>>>> with steepest descent to force of 2000 and constraint=none, the system
>>>>>>> converged in 754 steps with positive potential energy. In subsequent
>>>>>>> simulated annealing with constraint all bonds it starts giving link
>>>>>>> warning in 0 step with rms 7407.805164, max 66989.116545 (between atom
>>>>>>> 94 and 117) and a list of bond thar rotated more than 30 degree almost
>>>>>>> atom number belonging to chaps molecule.
>>>>>> You've set up a system that isn't stable, but we don't have enough
>>>>>> information to have any idea why. "I tried with new box size" doesn't
>>>>>> go
>>>>>> close to describing your method in enough detail for anyone to know
>>>>>> where
>>>>>> you went wrong.
>>>>>>
>>>>>> See http://www.gromacs.org/Documentation/Terminology/Blowing_Up for
>>>>>> generic
>>>>>> tips
>>>>>>
>>>>>> Mark
>>>>>>
>>>>>>> Please help.
>>>>>>> shahid Nayeem
>>>>>>>
>>>>>>> On Thu, Jan 27, 2011 at 7:06 PM, Justin A. Lemkul<jalemkul at vt.edu>
>>>>>>>  wrote:
>>>>>>>> shahid nayeem wrote:
>>>>>>>>> Dear All
>>>>>>>>>
>>>>>>>>> I am sending this mail again on user list because my reply to Mark’s
>>>>>>>>> query was not uploaded on the list.
>>>>>>>>>
>>>>>>>>> Original messge:
>>>>>>>>>
>>>>>>>>> I am trying to prepare a solvation box of chaps. After generating
>>>>>>>>> .itp
>>>>>>>>> and .gro at ProDrg and thorough check of charges, I started with a
>>>>>>>>> box
>>>>>>>>> size of 6x6x6. Energy minimization, simulated annealing (Cooling
>>>>>>>>> under
>>>>>>>>> high pressure and again heating at normal pressure) as well as final
>>>>>>>>> equilibration ran smoothly. But finally I get a box where all water
>>>>>>>>> molecules get accumulated in two three small region within the box
>>>>>>>>> and
>>>>>>>>> all chaps molecules gets accumulated in another small regions.I
>>>>>>>>> wanted
>>>>>>>>> near random uniform distribution of chaps in water. Any help from
>>>>>>>>> user, where I am wrong and what should I do.
>>>>>>>>>
>>>>>>>>> Reply to query.
>>>>>>>>>
>>>>>>>>> I created a box of 6x6x6 inserting 7 molecule of chaps with (genbox
>>>>>>>>> –ci 7 chaps.gro).Then I solvated the output box  with genbox using
>>>>>>>>> -maxsol 500 and spc216.gro. On visualization, at this stage itself
>>>>>>>>> uniform solvation did not occur (I got water in one region and chaps
>>>>>>>>> molecule in other region) but I observed a similar situation while
>>>>>>>> If your box was not completely solvated, then don't use -maxsol.  A
>>>>>>>> box
>>>>>>>> of
>>>>>>>> 6x6x6 nm should require more than 500 molecules of water to fill.  If
>>>>>>>> you're
>>>>>>>> trying to achieve some specific mole fraction or concentration, then
>>>>>>>> re-figure your box size.
>>>>>>>>
>>>>>>>> -Justin
>>>>>>>>
>>>>>>>>> preparing 10M urea salvation box. This was followed by 1ns simulated
>>>>>>>>> annealing from temp 300K to 0K and pressure 100 bar, then 1ns
>>>>>>>>> simulated annealing from temp. 0k to 300k and then ins
>>>>>>>>> equilibriation
>>>>>>>>> at this temperature. In case of urea finally I got uniformly
>>>>>>>>> solvated
>>>>>>>>> urea_water_box but in chaps I couldn’t get it.
>>>>>>>>>
>>>>>>>>> Shahid Nayeem
>>>>>>>> --
>>>>>>>> ========================================
>>>>>>>>
>>>>>>>> Justin A. Lemkul
>>>>>>>> Ph.D. Candidate
>>>>>>>> ICTAS Doctoral Scholar
>>>>>>>> MILES-IGERT Trainee
>>>>>>>> Department of Biochemistry
>>>>>>>> Virginia Tech
>>>>>>>> Blacksburg, VA
>>>>>>>> jalemkul[at]vt.edu | (540) 231-9080
>>>>>>>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>>>>>>>
>>>>>>>> ========================================
>>>>>>>> --
>>>>>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>> Please search the archive at
>>>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>>>>>>> Please don't post (un)subscribe requests to the list. Use the www
>>>>>>>> interface
>>>>>>>> or send it to gmx-users-request at gromacs.org.
>>>>>>>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>
>>>>>> --
>>>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>> Please search the archive at
>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>>>>> Please don't post (un)subscribe requests to the list. Use the www
>>>>>> interface
>>>>>> or send it to gmx-users-request at gromacs.org.
>>>>>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>>
>>>> --
>>>> ========================================
>>>>
>>>> Justin A. Lemkul
>>>> Ph.D. Candidate
>>>> ICTAS Doctoral Scholar
>>>> MILES-IGERT Trainee
>>>> Department of Biochemistry
>>>> Virginia Tech
>>>> Blacksburg, VA
>>>> jalemkul[at]vt.edu | (540) 231-9080
>>>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>>>
>>>> ========================================
>>>> --
>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>> Please search the archive at
>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>>> Please don't post (un)subscribe requests to the list. Use the www
>>>> interface
>>>> or send it to gmx-users-request at gromacs.org.
>>>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>
>> --
>> ========================================
>>
>> Justin A. Lemkul
>> Ph.D. Candidate
>> ICTAS Doctoral Scholar
>> MILES-IGERT Trainee
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>> ========================================
>>
> 

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

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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