Re: extracting waters from DCD

From: Marc Baaden (
Date: Wed Aug 09 2006 - 08:36:03 CDT

the Gromacs package provides such a tool. You'd do the following:

1) transform your dcd trajectory to trr with catdcd
2) use trajorder to order the water molecules in each timeframe
   according to their distance wrt to your point of interest
3) use trjconv to write a subtrajectory with the system and only the
   n (maybe 200 ?) first waters
4) use catdcd to transform the trr back to dcd

we have indeed used that to gain some space and preserve eg a trajectory
of a protein and its nearest solvation shell.

As Nicholas mentioned, you will of course loose the "identity" of the
water molecules. So if you were to look at things like residence times
you'd have a problem.

Marc Baaden

>>> Nicholas M Glykos said:
>> > I have a trajectory file for my simulation of a protein in a big water bo
>> > Now to reduce the size i can extract and writeout the trajectory for 'jus
>> > the protein' from this system by using "catdcd" but my aim is to extract
>> > some waters also (only those water molecules which are within 3.5 Ang. of
>> > the protein surface...for each frame).Any suggestions would be of great
>> > help.
>> Not easy: The number of water molecules around the solute (and within any
>> given cutoff, say 3.5A) will vary from frame to frame. What this implies
>> is that in the scenario you describe you can not have a constant-number-
>> of-atoms system.
>> I can imagine a solution where you decide to keep a constant number of
>> water molecules around the solute, and you don't care about the real
>> (PSF-wise) identity of individual molecules. As far as I know no such
>> program or script exists.
>> Nicholas

 Dr. Marc Baaden  - Institut de Biologie Physico-Chimique, Paris      -
 FAX: +33 15841 5026  -  Tel: +33 15841 5176  ou  +33 609 843217

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:27 CST