Re: Removing some water from a .dcd file or .coor file

From: Brian Novak (
Date: Sat Mar 27 2010 - 10:27:27 CDT

I think you would have to write an index file for each step for this to
work with catdcd if you wanted to keep waters based on distance, since
they change for each step. For visualization with VMD, you also have to
worry about keeping the number of waters constant between frames.

VMD can select only the protein and the waters near the protein (protein
or water and same residue as (pbwithin 3 of protein)), but as far as I
know there is no easy way to write extra waters so the number stays
constant for all frames. I have done this using matdcd to read the dcd
file, but it is slow since it doesn't do anything special to avoid some
of the distance calculations and it is MATLAB. I may try to change it
to Fortran or write a VMD script to do it.

If you don't care about keeping a trajectory that can be easily
visualized with VMD, then there is an example in the VMD 1.8.7 manual
(section 12.3) for writing each frame to a separate file.


On 3/18/2010 7:29 AM, wrote:
> Hi Rabab,
> to delete water molecules I use the tool catdcd combined with an
> indexfile containing all atoms for your 'new' .dcd file. The indexfile
> could be constructed with a vmd script.
> Best regards,
> Tillmann
> Quoting Rabab Toubar <>:
>> Hi all,
>> I have searched the mailing list and there are lots of suggestions
>> to totally remove the waters from a .dcd file after its generation. I
>> was wondering if there is a way to remove some of the waters
>> (specified by angstroms or solvation shells) to be removed so as to
>> downsize the solvation box.
>> Any suggestions?
>> Thanks
>> Rabab Toubar

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:55:37 CST