Hi,
On Thu, Mar 20, 2014 at 3:01 PM, <Juan.Aranda.uv.es> wrote:
>
> Where, donormask :2-395&.N*,O* is the enzyme donnor atoms
> and acceptormask :396-415&.N*,O* is the dna acceptor atoms, bridgeout
> still counts bridging waters between enzyme-water-enzyme, and dna-water-
> dna, apart from counting enzyme-water-dna.
>
This seems like the correct behavior. A solute - solvent hydrogen bond is
by definition an intermolecular hydrogen bond, so I hadn't considered
excluding U-V-U (where U is solute and V is solvent) from the bridging
calculation with 'nointramol'. There could potentially be a future option
to exclude these types of bridges.
> Is there a way to count only the waters involved in bridges between dna
> and enzyme (only enzyme-water-dna)?
>
No, see above. You could write a script in your favorite language to filter
out bridges you don't want (my fave is a combo of bash and awk).
>
> I've noticed that if you decrease the number of residues involved (i.e.
> :2-215&.N*,O* both for donnor and acceptor) the programme works.
> I guess it has to do with allocation of memory or something related...
>
That strongly indicates to me that this has something to do with memory.
How much memory is available to your system? The 'hbond' command can be
pretty memory-hungry. Cpptraj stores information on each found
solute-solute hydrogen bond, each solute-solvent hydrogen bond (with only
the solute part being uniquely identified), and each unique bridge (i.e.
unique combination of residues being bridged). This is on top of any other
memory being used by actions/topologies/data sets etc.
However, if you're only interested in the inter-molecular hydrogen bonds
for the second case, you could try adding the 'nointramol' keyword to that
as well, which will reduce the memory requirements somewhat.
Hope this helps,
-Dan
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Mar 20 2014 - 16:00:04 PDT
