[gmx-users] [ pairs_nb ] directive problems

[Mark Abraham]

Hi Joao,

That's great. You can find my guess at a fixed version, based on the 5.0
bugfix branch, at https://gerrit.gromacs.org/#/c/4666/. There's a download
link there, where you can get the putative fix in any format that suits
you, e.g. various tarballs. You can build and install that in the usual
way. Do let us know how you go, here or off-list as appropriate!

Mark

On Fri, May 29, 2015 at 11:19 AM João Martins <joaomartins139 at gmail.com>
wrote:

> Hi Mark,
>
> I can do that, where can I find either the fix or the fixed version?
>
> João
>
>
>
> *Joao Martins*
>
> joaomartins139 at gmail.com
>
> On Thu, May 28, 2015 at 4:11 PM, Mark Abraham <mark.j.abraham at gmail.com>
> wrote:
>
> > Hi,
> >
> > Thanks for the good description of the problem!
> >
> > It does look like [ pairs_nb ] is incompletely supported in grompp -
> that's
> > what that error message says. I would speculate that this has never
> worked
> > since Berk introduced it in 2008, to replace an old implementation of
> > [pairs] type 3. This can be fixed, but I don't have any inputs that
> > actually use the new code. Are you able to help check whether a fix works
> > as you expect, please?
> >
> > Mark
> >
> > On Thu, May 28, 2015 at 11:01 AM João Martins <joaomartins139 at gmail.com>
> > wrote:
> >
> > > Hi everyone,
> > >
> > > I'm trying to use the pairs_nb directive in order to decouple charge
> and
> > > VDW interactions of a small ligand in water but I'm running into
> problems
> > > with the implementation and understanding the manual. I know I could be
> > > using the couple-moltype and couple-intramol commands in my mdp file,
> but
> > > in my binding decoupling step those are not possible as I have my
> ligand
> > > and protein itps merged and as so I'm trying pairs_nb on a simpler
> > system.
> > >
> > > On the manual's table 5.5 the directive takes qi, qj (e); V**; W**.
> Using
> > > the combination rule 1 I've manually combined the LJ parameters for all
> > > pairs present in the ligand itp [ pairs ] directive and placed them in
> > the
> > > [ pairs_nb ] section like so:
> > >
> > > ...
> > > [ pairs ]
> > >
> > > [ pairs_nb ]
> > >     1     4     1  -0.3500 -0.2840 0.358141285 0.2343
> > >     1     7     1  -0.3500 -0.3910 0.358141285 0.2343
> > >     1    10     1  -0.3500 -0.3910 0.358141285 0.2343
> > >     1    18     1  -0.3500 0.0900 0.298451071 0.185231995
> > >     1    19     1  -0.3500 0.0900 0.298451071 0.185231995
> > >     2     5     1  -0.2840 -0.3500 0.358141285 0.2343
> > > ...
> > >
> > > One would believe this agrees with the manual, as it has, for line 1:
> > >
> > > atom number 1; atom number 2; function type; charge atom 1; charge atom
> > 2;
> > > combined sigma; combined epsilon
> > >
> > > However, I get this error when using it:
> > >
> > > Software inconsistency error:
> > > unknown directive
> > > For more information and tips for troubleshooting, please check the
> > GROMACS
> > > website at http://www.gromacs.org/Documentation/Errors
> > >
> > > I've been able before to at least force intramolecular VDW interactions
> > to
> > > be kept throughout the decoupling steps by using the [ nonbond_params ]
> > for
> > > both state A and state B atom types, but since my goal here was to also
> > > decouple instead of annihilating the intramolecular charge
> interactions,
> > > I'm playing around with the [ pairs_nb ] directive.
> > >
> > > Anyone with more experience than me using this can please tell me what
> I
> > am
> > > doing wrong?
> > >
> > > Best regards,
> > > João
> > >
> > >
> > > *Joao Martins*
> > >
> > > joaomartins139 at gmail.com
> > > --
> > > Gromacs Users mailing list
> > >
> > > * Please search the archive at
> > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > posting!
> > >
> > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > >
> > > * For (un)subscribe requests visit
> > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > > send a mail to gmx-users-request at gromacs.org.
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.