Hi Yunjie,
Note I am pretty sure that you cannot have different EEL and VDW cutoffs for
GB calculations in PMEMD. Also note that it is required that vdw_cutoff >=
es_cutoff. This is a function of the fact that such calculations are
expected to deal with the remaining electrostatic contributions in another
way, such as via a reciprocal calculation as in PME.
To do what you want to do I think you will have to get your hands dirty in
the code. Look in egb.f for 'cut' and you will see where the tests are etc.
You can then modify this to split out the VDW calculation separately and
treat it with a different cutoff value.
With regards to the PMEMD compilation problem I suspect it is linking in the
incorrect MPI libraries. Try editing the config.h file - blank out the
definition of MPI_LIBS and then change all instances of ifort to mpif90.
Also make sure that your copy of lam was compiled with this version of the
Intel compiler as well.
Good luck,
Ross
> -----Original Message-----
> From: amber-bounces.ambermd.org [mailto:amber-bounces.ambermd.org] On
> Behalf Of Yunjie Zhao
> Sent: Monday, May 04, 2009 5:58 AM
> To: AMBER Mailing List
> Subject: Re: [AMBER] problem with cutoff
>
> Hi Ross,
>
> Thanks for your reply. I am trying to do molecular simulations
> with different cut-offs of EEL / VDW in implicit solvent model. Maybe
> the
> cutoff should be much larger. It seems that the sander can not do this.
> I have tried to install PMEMD. But there is a problem. Can you tell
> me
> what the problem is?
>
> The input and output are as follows:
> $./configure linux_em64t ifort lam
> Intel ifort compiler found; version information:
> Version 10.1
> Use Intel MKL? (better performance) (answer yes or no):
> no
> File config_data/fft.pubfft being used...
> File config_data/linux_em64t.ifort being used...
> File config_data/interconnect.lam being used...
> configure assumes lam files are in /opt/lammpi.
> PMEMD Configurate successfully completed.
>
> $ make install
> cd src && make install
> make[1]: Entering directory `/export/home/zyj/amber9/src/pmemd/src'
> ifort -o pmemd gbl_constants.o gbl_datatypes.o state_info.o
> file_io_dat.o
> parallel_dat.o mdin_ctrl_dat.o mdin_ewald_dat.o prmtop_dat.o
> inpcrd_dat.o
> dynamics_dat.o img.o parallel.o pme_direct.o pme_recip.o pme_fft.o
> fft1d.o
> bspline.o pme_force.o pbc.o nb_pairlist.o cit.o dynamics.o bonds.o
> angles.o
> dihedrals.o runmd.o loadbal.o shake.o runmin.o constraints.o
> axis_optimize.o
> gb_ene.o veclib.o gb_force.o timers.o pmemd_lib.o runfiles.o file_io.o
> bintraj.o pmemd_clib.o pmemd.o random.o degcnt.o erfcfun.o nmr_calls.o
> nmr_lib.o get_cmdline.o master_setup.o alltasks_setup.o pme_setup.o
> ene_frc_splines.o nextprmtop_section.o -L/opt/lammpi/lib -llamf77mpi
> -lmpi -llam -ldl -lpthread -limf -lsvml
> -Wl,-rpath=/opt/lam/gnu/lib:/opt/pgi/linux86-64/7.0-
> 7/libso:/opt/lammpi/lib:/opt/intel/fce/10.1.012/lib
> /opt/lammpi/lib/liblam.so: undefined reference to `openpty'
> make[1]: *** [pmemd] Error 1
> make[1]: Leaving directory `/export/home/zyj/amber9/src/pmemd/src'
> make: *** [install] Error 2
>
> Thanks!
>
> 2009/5/4 Ross Walker <ross.rosswalker.co.uk>
>
> > Hi Yunjie,
> >
> > You cannot do this with Sander and I'm not sure why you'd want to. In
> fact
> > I'm not sure why you think an electrostatic cutoff in GB is
> warrantied at
> > all - especially one as small as 9 angstroms...
> >
> > What do you hope to gain by having a second even smaller vdw cutoff?
> You
> > still need to generate a pair list and loop over atoms so it is
> unlikely to
> > save you much if anything at all. PMEMD supports dual cutoffs for
> periodic
> > boundary simulations, where potentially it can make sense but I do
> not
> > think
> > it supports this for GB. You could hack it into egb.f yourself fairly
> > easily
> > but I don't see why you'd want to do that unless you wanted to
> experiment
> > with EEL / VDW effects as a function of cut off distance.
> >
> > In my opinion (others may differ here) if you are NOT doing periodic
> > boundary simulations with PME (such as GB) then you should NOT use
> any form
> > of cutoff... period!
> >
> > All the best
> > Ross
> >
> > > -----Original Message-----
> > > From: amber-bounces.ambermd.org [mailto:amber-bounces.ambermd.org]
> On
> > > Behalf Of Yunjie Zhao
> > > Sent: Sunday, May 03, 2009 8:53 PM
> > > To: AMBER.ambermd.org
> > > Subject: [AMBER] problem with cutoff
> > >
> > > Dear Amber users,
> > >
> > > Using Amber 9, I am trying to do molecular simulations with GB
> > > implicit
> > > solvent model. I have done a cut-off to distinguish short-range and
> > > long-range interactions. Now I am trying to distinguish the vdw and
> > > electrostatic interactions, such as cutoff of vdw with 6 angstrom
> and
> > > cutoff of electrostatic with 9 angstrom. How can I do this? Can
> anyone
> > > tell
> > > me how to write the input file?
> > >
> > > My input file:
> > > MD Generalised Born simulation
> > > &cntrl
> > > imin = 0, ntb =0,
> > > igb = 1, ntpr = 10000, ntwx = 10000,
> > > ntt = 3,gamma_ln = 1.0,
> > > tempi = 300.0, temp0 = 300.0,
> > > nstlim = 10000000, dt = 0.001,
> > > cut = 9
> > > /
> > >
> > > Thanks!
> > > _______________________________________________
> > > 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
> >
> _______________________________________________
> 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 May 20 2009 - 14:33:03 PDT
