Dear Prof. Case,
Another interesting observation was when I applied a boost potential using
coefficients as low as b1, b2 = 0.001.
The calculation forwards usually until no boost is applied because the
unboosted energy is still higher than the threshold energy (-7438 kcal/mol)
but the moment it becomes lower and boost is applied to the system,
very high energy values are obtained in the next step. A part of amd.log
file.
100 4900 -6700.099418458845 221.870281367563
1.000000000000 1.000000000000 0.000000000000
0.000000000000
100 5000 -6197.826899794247 220.560035056435
1.000000000000 1.000000000000 0.000000000000
0.000000000000
100 5100 -6583.882674558405 226.478496809490
1.000000000000 1.000000000000 0.000000000000
0.000000000000
100 5200 -6956.412920554485 225.212923988700
1.000000000000 1.000000000000 0.000000000000
0.000000000000
100 5300 -7166.396857408192 213.378022024408
1.000000000000 1.000000000000 0.000000000000
0.000000000000
*100 5400 -7439.902389557417 218.166992922314
0.636276736925 1.000000000000 0.384911304309
0.000000000000*
* 100 5500 17083603.195437256247 460.907964122482
1.000000000000 1.000000000000 0.000000000000
0.000000000000 100 5600**********************
967.790122159757 0.000000000000
1.000000000000********************** 0.000000000000*
What could be the reason?
Thank you,
Chetna
On Fri, Nov 16, 2018 at 11:40 AM Chetna Tyagi <cheta231.gmail.com> wrote:
> Dear Prof. Case,
>
> Many thanks for your suggestions.
> I ran a short simulation with iwrap =0 with ntpr =1 and what I noticed is
> that at 175th step the potential energy value is -5729 kcal/mol and
> suddenly shoots to 43796 kcal/mol by the next step.
>
> NSTEP = 175 TIME(PS) = 2506000.363 TEMP(K) = 456.79 PRESS
> = 0.0
> Etot = 4365.3973 EKtot = 10094.7480 *EPtot =
> -5729.3507*
> BOND = 2780.9813 ANGLE = 2895.3560 DIHED =
> 265.1213
> 1-4 NB = 369.9476 1-4 EEL = 875.2511 VDWAALS =
> -11459.4185
> EELEC = -1456.5895 EHBOND = 0.0000 RESTRAINT =
> 0.0000
> EAMD_BOOST = 0.0000
>
> ------------------------------------------------------------------------------
>
>
> NSTEP = 176 TIME(PS) = 2506000.365 TEMP(K) =********* PRESS
> = 0.0
> Etot = ************** EKtot = ************** * EPtot =
> 43796.3823*
> BOND = 1868.3324 ANGLE = 2993.9430 DIHED =
> 265.9292
> 1-4 NB = 522.2288 1-4 EEL = 876.3629 VDWAALS =
> 38681.8646
> EELEC = -1412.2787 EHBOND = 0.0000 RESTRAINT =
> 0.0000
> EAMD_BOOST = 0.0000
>
> ------------------------------------------------------------------------------
>
> I am using very low b1 and b2 coefficients (0.10, 0.08) to calculate
> accelerated MD parameters. I had even tried creating a bigger boy of CHCl3
> with peptide in it and still the same results.
>
> Regarding previous preparation calculations like equilibrium etc were
> carried out fine with no abnormal values or results. iwrap=1 option was
> used for all steps.
>
> I don't know what are downstream problems.
>
> Please advice as to how can this be solved?
>
> Thank you,
> Chetna
>
>
> On Thu, Nov 15, 2018 at 10:30 PM David Case <david.case.rutgers.edu>
> wrote:
>
>> On Thu, Nov 15, 2018, Chetna Tyagi wrote:
>> >
>> >Upon carrying out accelerated MD on a peptide solvated in water and
>> >chloroform solvents in *Amber14 pmemd.CUDA.MPI*, I realized that when
>> boost
>> >to the potential is applied to the Chloroform system, the various values
>> in
>> >the output result in asterisks....
>> >
>> >For e.g. NSTEP = 1000 TIME(PS) = 2506002.013 TEMP(K) =*********
>> >
>> >Based on previous answers, the problem can be traced to high wrapping
>> >values which was,
>> >
>> >*wrapping first mol.: 136974.67863 1318602.52880 -1118357.40723*
>> >I used *iwrap =1* option in both cases.
>>
>> The bad wrapping numbers look more like a downstream problem, caused by
>> some earlier problem; but I could be wrong. What happens if you run a
>> short simulation (since the problem appears already at step 1000),
>> setting iwrap=0? You might also print much more often (i.e. set ntpr=1)
>> to get more clues about what might be going wrong.
>>
>> ....good luck...dac
>>
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>
>
> --
> Best wishes
> Chetna
>
>
--
Best wishes
Chetna
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Nov 16 2018 - 03:30:02 PST
