AMBER Archive (2007)Subject: RE: AMBER: thermodynamic integration
From: Cooper, Matthew (matthew_cooper_at_merck.com)
Date: Fri Nov 09 2007 - 08:03:32 CST
For the jobs, which remove the charge on a solvated molecule to measure
the free energy of charging, there are two types of input decks:
For the first part:
Imin = 0, irest = 0,
Nstlim = 100000,
Ntx = 1, ntb = 2, ntp = 1, ntt = 1,
Tautp=1, tauup=5, temp0=298.0,
Ntave=1250, dt=0.002,
Ntc =2, ntf = 2, cut = 9.,
Ntpr = 125, ntwe = 125, ntwx = 125, iwrap = 1,
Scee = 1.2, icfe = 1,
Clambda = 0.0,
For the next 5 parts (where clambda equals 0.2, 0.4, 0.6, 0.8 and 1.0
respectively and the two states are charged and uncharged)
Imin = 0, irest = 1,
Nstlim = 100000,
Ntx = 5, ntb = 2, ntp = 1, ntt = 1,
Tautp=1, tauup=5, temp0=298.0,
Ntave=1250, dt=0.002,
Ntc =2, ntf = 2, cut = 9.,
Ntpr = 125, ntwe = 125, ntwx = 125, iwrap = 1,
Scee = 1.2, icfe = 1,
Clambda = 0.2,
For now, let us forget that, on the exact same molecules, sometimes it
works and sometimes it does not because I think that the two problems
may be in play.
What I meant by "running two molecules with identical starting positions
but different charges, one may fail but the other will not" was that I
take an ffd files, and then charge it with different, though similar,
charges. I then submit these two identical except for their charges
molecules with the same input decks to sander. One set of charges will
result in DV/DL values while the other will not. This usually does not
happen but occasionally does.
One molecule is 12_diacetoxyethane. It runs fines under one set of
charges but when those charges are changed the run may fail.
Interestingly, running a molecules that failed in AMBER 9 (no DV/DL
values) with the same input decks but using AMBER 7 results in DV/DL
values.
Thanks
Matthew
>
> I am experiencing difficulties with thermodynamic integration in Amber
> 9.
>
> I use the same input deck for numerous integrations. Occasionally,
there
> will be no data for DV/DL, though a summary will be printed that
> contains only zeros.
We need more details about what happens when you get all zeros. What
values
of clambda and klambda and icfe are you using?
>
> Rerunning the script will sometimes result in DV/DL averages being
> calculated though not always. If I run two molecules with identical
> starting positions but different charges, one may fail while the other
> will not.
This is not so clear. Do you mean that running the *exact* same job
gives
different answers at different times? Also, I don't really understand
what
you mean by "running two molecules with identical starting positions but
different charges, one may fail but the other will not". What exactly
is
the "one" that fails and "the other" that does not?
...dac
------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, contains
information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station,
New Jersey, USA 08889), and/or its affiliates (which may be known
outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD
and in Japan, as Banyu - direct contact information for affiliates is
available at http://www.merck.com/contact/contacts.html) that may be
confidential, proprietary copyrighted and/or legally privileged. It is
intended solely for the use of the individual or entity named on this
message. If you are not the intended recipient, and have received this
message in error, please notify us immediately by reply e-mail and then
delete it from your system.
------------------------------------------------------------------------------
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
|