[gmx-users] Exclusion - determine neighboring atoms by atom indices or by bond connections?

[Mark Abraham]

Hi,

On Wed, 26 Jul 2017 22:53 lan hoa Trinh <trinhlanhoa at gmail.com> wrote:

> Dear Gromacs community,
> In the Gromacs manual, if nrexcl = 3, that means atoms i+1, i+2, i+3 are
> excluded from non-bonded interactions with atom i. I don't know how Gromacs
> determine the neighboring atoms, based on atom indices or bond connections?
>

Every mention of nrexcl in the reference manual specifically mentions that
it refers bonded connectivity. Atom indices couldn't work. What were you
reading?

For example, if I have a simple connection between CA and CB atoms like the
> following
>
> CA(1) -------------CA(3)---------------CA(5)----------------CA(7)
> |                       |                         |                     |
> CB(2)              CB(4)                CB(6)               CB(8)
>
> Indices of atoms are shown in the brackets. If the neighbouring atoms are
> determined based on atom indices, that means CB(2) will not form non-bonded
> interactions with CA(3), CB(4),CA(5) but still form non-bonded interactions
> with CA(1), that is what I don't want since CB(2) and CA(1) already form
> bonded interaction (harmonic bond energy)
>

The implementation works to exclude nonbonded interactions with bonded
neighbours in all directions. The most convenient way of expressing and
implementing it is to loop over all atoms with index i and exclude all
bonded neighbours whose atom indices are greater than i. Excluding the
interaction between i and atoms with smaller indices were handled when
considering those atoms.

Mark


It really confused me. I am looking forward to hearing from you.
> Thank you very much in advance.
> Best regards,
> *Lan Hoa*
> --
> 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.
>