Re: use of tiElecLambdaStart (NAMD 2.7b1)

From: Chris Harrison (char_at_ks.uiuc.edu)
Date: Fri Mar 27 2009 - 17:28:22 CDT

Oops, sorry, forgot to mention tiVdwShiftCoeff. My mistake. The default is 5 and as Jerome said would also have to be changed to effectively "turn off" the soft-core. So for those wanting to run TI perturbations without softcore:

tiVdwShiftCoeff 0
tiElectLambdaStart 0
tiVdwLambdaEnd 1

For those now asking "should I turn off soft core?" .... if you have to ask, it's probably best to leave soft-core "on" for your simulation. Further info on soft-core potentials can be found by reading the Beutler and Zacharias articles cited in the Alchemical Transformations sections of the NAMD2.7b1 manual.

C.

--
Chris Harrison, Ph.D.
Theoretical and Computational Biophysics Group
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
char_at_ks.uiuc.edu                          Voice: 217-244-1733
http://www.ks.uiuc.edu/~char              Fax:   217-244-6078
Jerome Henin <jhenin_at_cmm.chem.upenn.edu> writes:
> From: Jerome Henin <jhenin_at_cmm.chem.upenn.edu>
> Date: Fri, 27 Mar 2009 17:32:31 -0400
> To: Chris Harrison <char_at_ks.uiuc.edu>
> Cc: brmorgan_at_clarku.edu, namd-l_at_ks.uiuc.edu
> Subject: Re: namd-l: use of tiElecLambdaStart (NAMD 2.7b1)
> Return-Path: char_at_halifax.ks.uiuc.edu
> Message-ID: <a61b1c250903271432g3d236f2fne298bd40265d145c_at_mail.gmail.com>
> X-Spam-Status: No, score=-1.0 required=5.0 tests=AWL,BAYES_00,
> 	DNS_FROM_RFC_POST,SPF_PASS autolearn=no version=3.1.7-0+tcb1
> 
> Hi guys,
> Just a little remark about what Chris wrote: setting tiElecLambdaStart
> to 0 and tiVdwLambdaEnd to 1 will not remove the soft-core potential
> (except if the vdwShift parameter is set to 0), but it will result in
> "naked charges" at lambda close to 0 (resp. 1 for disappearing atoms),
> which typically cause big-time numerical issues in the simulation.
> Jerome
> 
> On Fri, Mar 27, 2009 at 3:33 PM, Chris Harrison <char_at_ks.uiuc.edu> wrote:
> > Brittany,
> >
> > Can you explain why you can't use a perturbation that maintains
> > neutrality?  While not balancing the charges may not lead to problems,
> > it is in general very bad form as you'll essentially be propogating to
> > an infinite charge via PME periodicity.
> >
> > The default value is 0.5.  Using a zero value with tiElecLambdaStart and a tiVdwLambdaEnd value of 1 will essentially remove the soft-core potential. Unless there is a specific reason to change tiElecLambdaStart it's probably best to leave it at its default of 0.5 .... is there a reason you want to change it?
> >
> > C.
> >
> >
> >
> > --
> > Chris Harrison, Ph.D.
> > Theoretical and Computational Biophysics Group
> > NIH Resource for Macromolecular Modeling and Bioinformatics
> > Beckman Institute for Advanced Science and Technology
> > University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
> >
> > char_at_ks.uiuc.edu                          Voice: 217-244-1733
> > http://www.ks.uiuc.edu/~char              Fax:   217-244-6078
> >
> >
> > brmorgan_at_clarku.edu writes:
> >> From: brmorgan_at_clarku.edu
> >> Date: Fri, 27 Mar 2009 12:00:51 -0400 (EDT)
> >> To: namd-l_at_ks.uiuc.edu
> >> Subject: namd-l: use of tiElecLambdaStart (NAMD 2.7b1)
> >> Return-Path: char_at_halifax.ks.uiuc.edu
> >> Message-ID: <51618.146.189.58.124.1238169651.squirrel_at_cmail.clarku.edu>
> >> X-Spam-Status: No, score=-0.5 required=5.0 tests=AWL,BAYES_00,NO_REAL_NAME
> >>       autolearn=no version=3.1.7-0+tcb1
> >>
> >> Hi,
> >>
> >> In the thermodynamic integration implemented in NAMD 2.7b1, should
> >> tiElecLambdaStart always be used, and how would one determine the best
> >> value to use? How much does the use of a non-zero tiElecLambdaStart affect
> >> the results? My system changes in total charge as a function of lambda and
> >> I plan to use PME.
> >>
> >> Thank you,
> >> Brittany Morgan
> >>
> >
> >
> 

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:40 CST