Re: NNB in PSF and alchemify

From: Jerome Henin (jhenin_at_cmm.chem.upenn.edu)
Date: Tue Jun 26 2007 - 13:46:50 CDT

Hi Luca,
It is a rather strange format, and it took me a long time to figure it
out, but by now I think that part of the code is well-tested.
Here is roughly how it works: the section begins with a series of
lists of excluded atoms, and those lists are then assigned to other
atoms by a list of offsets within that first list... Since it is
probably very unclear, let me use an example: if there are 8 atoms in
total and I want to exclude {1, 2} from 3 and {1, 4, 5} from 6, here
is the NNB section I need:

5 !NNB
1 2 1 4 5
0 0 2 2 2 5 5 5

In the case of alchemify, since it is always about excluding set A
from set B, the section always starts with indices of set A, repeated
as many times as set B has atoms. The rest of the section indirectly
says what atoms belong to set B. Not that a different file, excluding
set B from set A, would have the same physical results.

I hope it makes sense...
Jerome

On 6/26/07, Luca Bellucci <bellucci14_at_unisi.it> wrote:
> Hi all,
> I should clarify NNB (non-bonded interaction ) section when I create psf with
> alchemify program.
> In the NNB section of new psf file are present only atoms with lambda values
> of 1 in file.fep .
> Why not there are atoms with lambda equal to -1 ?
> These atoms do not interact each other but in NNB section they are only of
> one type.
> It is wrong?
> Which is the format of NNB in PSF file?
> example:
> PDB or Fep file:
>
> ATOM 19 HB2A STA 2 40.764 46.498 10.639 1.00 -1.00
> ATOM 20 OGA STA 2 42.181 47.086 9.211 1.00 -1.00
> ATOM 21 HG1A STA 2 42.884 46.570 8.738 1.00 -1.00
> ATOM 22 CBB STA 2 41.325 46.085 9.771 1.00 1.00
> ATOM 23 HB1B STA 2 41.965 45.250 10.138 1.00 1.00
> ATOM 24 HB2B STA 2 40.764 46.498 10.639 1.00 1.00
> ATOM 25 ..
> ATOM 26..
>
>
> PSF file:
> 30 !NNB
> 22 23 24 25 26 27 22 23
> 24 25 26 27 22 23 24 25
> 26 27 22 23 24 25 26 27
> 22 23 24 25 26 27
>
> Thanks in advance
> Luca
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:33 CST