Re: removing waters from big dcd files

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Mon Jul 16 2007 - 15:13:20 CDT

Hi Rebecca,
these are actually serial numbers, not indices (which start from 0), and
are likely giving you a dcd file that doesn't correspond to the atoms
you want. I'd recommend getting the indices as follows in vmd:

set sel [atomselect top protein]
set ofile [open "protein.ind" "w"]
puts $ofile [$sel get index]
close $ofile

This should give catdcd what it wants.

Best,
Peter

regafan_at_usc.es wrote:
>
> Hello,
> Thanks a lot for your help. I get the indexes of the protein using
>
> seq ?f ?%04g? ?s " " 2573
>
>
> so I get a file of the type
>
> 0001 0002 0003 0004 (...)2573
>
> I have also tried with a index file of the type
>
> 1 2 3 (...) 2573
>
>
> But any of them gives me an adequate visualization of the dcd file
> generated with catdcd using vmd. Do you think is it problem of the
> index numbers?
>
>
>
> Rebeca Garcia Fandiņo
> Parc Cientific de Barcelona
> regafan_at_usc.es
>
>
>
>
> Estase citando Jerome Henin <jhenin_at_cmm.chem.upenn.edu>:
>
>> Hi Rebeca,
>>
>> On 7/16/07, regafan_at_usc.es <regafan_at_usc.es> wrote:
>>> Hello,
>>>
>>> I would like to remove the waters from from a dcd file obtained with
>>> NAMD/Charmm, and the problem is that the size of the file is 2.9 G
>>> (with waters). I have tried to use ptraj to obtain the "dry
>>> trajectory", and I got a quite strange trajectory, with a first and a
>>> last step different from the structure of the corresponding first and
>>> last .coor files. I have also tried to use catdcd, choosing the
>>> indexes of only the protein,
>>
>> I would have expected this to work. A problem you may have is the
>> difference between PDB files and VMD for atom numbering - VMD uses
>> 0-based indices while PDB indices are 1-based (that is what catdcd
>> expects). If you use VMD to determine the indices, add one to get PDB
>> indices.
>>
>> Jerome
>>
>>> but when I visualize the resulting
>>> trajectory, loading first the psf of the protein without the waters, I
>>> cannot visualize the tertiary sctructure, it is only a ribbon what I
>>> can see.
>>> I have seen that there is also a script for big files, "bigdcd", but I
>>> donīt know how to use it to eliminate the waters, and if this is
>>> posible.
>>> Could anyone help me with the treatment of this big dcd files?
>>> Thanks a lot!
>>>
>>> Rebeca Garcia Fandiņo
>>> Parc Cientific de Barcelona
>>> regafan_at_usc.es
>>>
>>>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:38 CST