In that input I just didn't want to dump anything, I was not doing
umbrella sampling.
If you want, just ignore those two files, and use these ones, which
are made for umbrella sampling:
Input file:
&cntrl
imin = 0,
ntx = 5, irest = 1,
ntpr = 5000, ntwr = 5000, ntwx = 10000,
ioutfm = 1,
iwrap = 1,
ntc = 2, ntf = 2,
ntb = 1, cut = 10.0,
ntt = 3, temp0 = 298.0, gamma_ln = 2.0, ig = -1,
nmropt = 1,
nstlim= 500000, dt = 0.002,
/
&wt type='DUMPFREQ', istep1=100 /
&wt type='END' /
DISANG=dist.RST_75
DUMPAVE=dist_vs_t_75
Restraint file:
# distance restraint
# move atom 19560 (a bicarbonate ion) across the channel, from r=90 to r=4
&rst
iat=19560,8098 r1=0.1, r2=75, r3=75, r4=125, rk2 = 20.0, rk3 = 20.0, ialtd=0,
&end
Ignacio
On Wed, Oct 17, 2012 at 10:13 AM, Elisa Frezza <elisa.frezza.gmail.com> wrote:
> Dear Ignacio,
>
> why do you not write &wt type='DUMPFREQ'?
>
> ELisa
>
> 2012/10/17 Elisa Frezza <elisa.frezza.gmail.com>
>
>> Yes,
>>
>> when I use ntr or other options I have no problem with pmemd.cuda.
>>
>> Now I'll try.
>>
>> Thank you
>>
>> ELisa
>>
>> 2012/10/17 Ignacio J. General <ijgeneral.gmail.com>
>>
>>> Elisa, I don't see any clear problem in your input, but sometimes even
>>> small details, like one extra or missing space in the input can cause
>>> problems. In particular, I remember that I had a problem once with a
>>> missing space in the restraint file (DISANG file).
>>>
>>> Here I copy a pair of input and nmropt restraint files that works for
>>> me, so you can try to copy the format. Hopefully, that will solve the
>>> problem.
>>>
>>> Input file:
>>>
>>> &cntrl
>>> imin = 0,
>>> irest = 0,
>>> ntx = 1,
>>> ntb = 1,
>>> cut = 10,
>>> ntp = 0, pres0 = 1.0, taup = 2.0,
>>> ntt = 3, tempi = 10, temp0 = 298,
>>> gamma_ln = 2.0, ig = -1,
>>> ntc = 2, ntf = 2,
>>> nstlim = 50000, dt = 0.002,
>>> ntpr = 5000, ntwx = 5000, ntwr = 5000,
>>> ioutfm = 1,
>>> nmropt = 1,
>>> /
>>> &wt type='END', /
>>> DISANG=restraint-distance
>>>
>>> Restraint file:
>>>
>>> # Restraint on unless bicarbonate distance to two atoms in protein is
>>> less than 3A.
>>> # 1230 BCA C 631 ALA C 5
>>> &rst
>>> ixpk= 0, nxpk= 0, iat= 19561, 10245, r1= 1.00, r2= 1.10, r3=3.00,
>>> r4=3.50,
>>> rk2=5.0, rk3=5.0, ir6=1, ialtd=0,
>>> &end
>>> #
>>> # 1230 BCA C 31 ALA CB 5
>>> &rst
>>> ixpk= 0, nxpk= 0, iat= 19560, 534, r1= 1.00, r2= 1.10, r3=3.00,
>>> r4=3.50,
>>> &end
>>>
>>> Also, I imagine that you have a working installation of pmemd.cuda,
>>> right? Does it work fine with other inputs?
>>>
>>> Ignacio
>>>
>>>
>>> On Wed, Oct 17, 2012 at 9:16 AM, Elisa Frezza <elisa.frezza.gmail.com>
>>> wrote:
>>> > Dear Ignacio,
>>> >
>>> > I tried to use only nmropt=1, but I have still the same problem.
>>> > The restraint file is:
>>> >
>>> > &rst iat=243,1001,
>>> > ninc=0,ir6=0,ifntyp=0,nstep1=1,nstep2=90000,
>>> > r1=0.0000E+00,r2=28.0000,r3=28.0000,r4=50.000,rk2=5,rk3=5,ir6=0/
>>> >
>>> > and the output related to this part is:
>>> >
>>> > Begin reading energy term weight changes/NMR restraints
>>> > WEIGHT CHANGES:
>>> > DUMPFREQ 1 0 0.000000 0.000000 0 0
>>> > ** No weight changes given **
>>> >
>>> > RESTRAINTS:
>>> > Requested file redirections:
>>> > DISANG = dist_chi.RST
>>> > DUMPAVE = restraint.log
>>> > Restraints will be read from file: dist_chi.RST
>>> > Here are comments from the DISANG input file:
>>> >
>>> > Number of restraints read = 1
>>> >
>>> > Done reading weight changes/NMR restraints
>>> >
>>> > Thanks you in advance
>>> >
>>> > ELisa
>>> >
>>> > 2012/10/17 Ignacio J. General <ijgeneral.gmail.com>
>>> >
>>> >> Dear Elisa, I have been experimenting with umbrella sampling lately,
>>> >> using GPUs, and I found that it's not possible to use pmemd.cuda with
>>> >> nmropt=1 and ntr=1 at the same time. It just stops, without any clear
>>> >> error message. But this was solved as soon as I changed to use sander
>>> >> (I don't remember now, but maybe it also works fine with pmemd).
>>> >>
>>> >> So, I think your choices are:
>>> >> 1) use sander (or, maybe, pmemd), or
>>> >> 2) use GPUs, but you should try to figure out a way to combine your
>>> >> restraints so that you end up using only nmropt or ntr, but not both.
>>> >>
>>> >> Regards,
>>> >>
>>> >> Ignacio
>>> >>
>>> >> On Wed, Oct 17, 2012 at 5:32 AM, Elisa Frezza <elisa.frezza.gmail.com>
>>> >> wrote:
>>> >> > Dear All,
>>> >> >
>>> >> > I am trying to do umbrella sampling, but the calculation does not
>>> run.
>>> >> > I use the following input:
>>> >> >
>>> >> > run: step 1 MD
>>> >> > &cntrl
>>> >> > imin = 0,
>>> >> > irest = 1,
>>> >> > ntx = 5,
>>> >> > nstlim =100000, dt = 0.002,
>>> >> > ntc = 2 , ntf = 2,
>>> >> > ntt = 3, gamma_ln = 2.0,
>>> >> > ntp=1, taup=1,
>>> >> > pres0=1,
>>> >> > tempi = 300.0, temp0 =300.0,
>>> >> > ntb = 2 ,
>>> >> > ntpr = 250 , ntwx = 250, ntwr = 500 ,
>>> >> > cut = 9,
>>> >> > nmropt=1,
>>> >> > restraintmask = :1-48,
>>> >> > restraint_wt=0.0005
>>> >> > /
>>> >> > &wt type='DUMPFREQ', istep1=1 /
>>> >> > &wt type='END' /
>>> >> > DISANG=dist_chi.RST
>>> >> > DUMPAVE=restraint.log
>>> >> >
>>> >> > and the following command:
>>> >> >
>>> >> > cmd="pmemd.cuda.MPI -O -i step11.in -o dimero_step11.out -p
>>> >> > dimero_spce.prmtop -c dimero_step10.rst -r dimero_s
>>> >> > tep11.rst -x dimero_step11.mdcrd -ref dimero_step10.rst"
>>> >> >
>>> >> > In the output file there is not problem, but the calculation is
>>> however
>>> >> > aborted:
>>> >> > forrtl: severe (174): SIGSEGV, segmentation fault occurred
>>> >> > Image PC Routine Line
>>> >> Source
>>> >> >
>>> >> > pmemd.cuda.MPI 000000000053FDB0 Unknown Unknown
>>> >> Unknown
>>> >> > pmemd.cuda.MPI 00000000004FF822 Unknown Unknown
>>> >> Unknown
>>> >> > pmemd.cuda.MPI 0000000000433F9E Unknown Unknown
>>> >> Unknown
>>> >> > pmemd.cuda.MPI 0000000000510C80 Unknown Unknown
>>> >> Unknown
>>> >> > pmemd.cuda.MPI 00000000004ECA8B Unknown Unknown
>>> >> Unknown
>>> >> > pmemd.cuda.MPI 000000000040539C Unknown Unknown
>>> >> Unknown
>>> >> > libc.so.6 000000315E01D994 Unknown Unknown
>>> >> Unknown
>>> >> > pmemd.cuda.MPI 00000000004052A9 Unknown Unknown
>>> >> Unknown
>>> >> >
>>> >> >
>>> >> > Can someone help me?
>>> >> >
>>> >> > Thank you in advance
>>> >> >
>>> >> > ELisa
>>> >> >
>>> >> >
>>> >> > --
>>> >> > Elisa Frezza
>>> >> > Ph.D. Student in Materials Science and Engineering
>>> >> > Dipartimento di Scienze Chimiche
>>> >> > Università di Padova
>>> >> > via Marzolo, 1
>>> >> > 35131 Padova - Italy
>>> >> > Phone: +39 049 827 5149
>>> >> > Skype: elisa.frezza
>>> >> > Emai: elisa.frezza.gmail.com
>>> >> > elisa.frezza.studenti.unipd.it
>>> >> > _______________________________________________
>>> >> > AMBER mailing list
>>> >> > AMBER.ambermd.org
>>> >> > http://lists.ambermd.org/mailman/listinfo/amber
>>> >>
>>> >> _______________________________________________
>>> >> AMBER mailing list
>>> >> AMBER.ambermd.org
>>> >> http://lists.ambermd.org/mailman/listinfo/amber
>>> >>
>>> >
>>> >
>>> >
>>> > --
>>> > Elisa Frezza
>>> > Ph.D. Student in Materials Science and Engineering
>>> > Dipartimento di Scienze Chimiche
>>> > Università di Padova
>>> > via Marzolo, 1
>>> > 35131 Padova - Italy
>>> > Phone: +39 049 827 5149
>>> > Skype: elisa.frezza
>>> > Emai: elisa.frezza.gmail.com
>>> > elisa.frezza.studenti.unipd.it
>>> > _______________________________________________
>>> > AMBER mailing list
>>> > AMBER.ambermd.org
>>> > http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>
>>
>>
>> --
>> Elisa Frezza
>> Ph.D. Student in Materials Science and Engineering
>> Dipartimento di Scienze Chimiche
>> Università di Padova
>> via Marzolo, 1
>> 35131 Padova - Italy
>> Phone: +39 049 827 5149
>> Skype: elisa.frezza
>> Emai: elisa.frezza.gmail.com
>> elisa.frezza.studenti.unipd.it
>>
>>
>>
>>
>
>
> --
> Elisa Frezza
> Ph.D. Student in Materials Science and Engineering
> Dipartimento di Scienze Chimiche
> Università di Padova
> via Marzolo, 1
> 35131 Padova - Italy
> Phone: +39 049 827 5149
> Skype: elisa.frezza
> Emai: elisa.frezza.gmail.com
> elisa.frezza.studenti.unipd.it
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Oct 17 2012 - 07:30:10 PDT
