[gmx-users] Writing initial coordinates to trajectory files

[Berk Hess]

>From: Brian Helfrich <brian at nanorex.com>
>Reply-To: helfrich at nanorex.com,Discussion list for GROMACS users 
><gmx-users at gromacs.org>
>To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>Subject: Re: [gmx-users] Writing initial coordinates to trajectory files
>Date: Fri, 30 Nov 2007 12:55:20 -0500
>
>Berk Hess wrote:
>>The initial coordinates are written immediately to the trr file,
>>although the file system might buffer data, you so could have to wait
>>until some more frames have been written.
>
>Thanks, I see that now. It seems that somewhere between grompp and mdrun's 
>initial trajectory output, my initial coordinates did change ever so 
>slightly:
>
>after_em.gro:
>    1LYSH     N    1   2.105   0.744   2.070
>    1LYSH    H1    2   2.056   0.668   2.027
>    1LYSH    H2    3   2.190   0.761   2.021
>
>full.trr:
>   natoms=      9035  step=         0  time=0.0000000e+00  lambda=         
>0
>   ...
>      x[    0]={ 2.10518e+00,  7.43687e-01,  2.07004e+00}
>      x[    1]={ 2.05600e+00,  6.68002e-01,  2.02700e+00}
>      x[    2]={ 2.19043e+00,  7.61087e-01,  2.02075e+00}
>
>But step 0 coordinates are output as you said, thanks.
>--Brian.
>
>--
>Brian Helfrich
>www.nanorex.com

A gro file does not have full accuracy.
Probably mdrun changed the initial coordinates because of the constraints.
The rounded coordinates in the gro file do not obey the constraints exactly.

Berk.

_________________________________________________________________
Talk with your online friends with Messenger 
http://www.join.msn.com/messenger/overview