[gmx-users] Re: Re: Probable error in trjcat; skipping frames

[Justin A. Lemkul]

Lipi Thukral wrote:
> Dear David,
> Dear Justin,
> 
> I have now installed the latest version of Gromacs 4.5.4. The problem 
> still persists. 
> The two files to concatenate do not have exactly the same timeframes. 
> However they do have something like the following:
> 
> File 1: 2091000.0000000
> File 2: 2090170.0000000
> 

Are these the starting times?

> May be I don't understand but its clear that the two files do not have 
> the same time-frames so why is trjcat skipping the frames?
> 

It's not skipping frames, it's over-writing frames in common.  If File2 starts 
at 2090170 and continues past 2091000, it overlaps with File1, and per trjcat -h:

"In case of double time frames the one in the later file is used."

So you're getting a fragmented trajectory.  You can concatenate them and keep 
double frames with trjcat -cat, but it's not yet clear to me why you're doing 
what you're doing, but it seems that trjcat is behaving exactly as advertised.

I would still be interested in seeing the full gmxcheck output (without 
copyright and help info stuff), not just how many frames are in each file, if 
you are still convinced of a problem.

-Justin

> Regards,
> Lipi
> 
> 
> On 4/27/2011 3:51 PM, Lipi Thukral wrote:
>>  Hi Justin,
>>
>>  I am using Gromacs 4.5 beta-2.
>>  gmxcheck with file 1 and file 2 tells me exactly the number of frames
>>  I specified.
>>  gmxcheck -f File1.xtc
>>  Frames: 2175
>>
>>  gmxcheck -f File2.xtc
>>  Frames: 1232
>>
>>  While after concatenating, it skips the frames and the combined
>>  trajectory results in 1266 frames.
>>  Do you attribute this to Gromacs4.5beta2 version? Is there a possible
>>  bug in that version for trjcat?
> 
> Maybe - but it's not worth finding out whether a bug existed then.
> There's no value in trouble-shooting a beta version that is known to
> have other problems. Get the latest GROMACS version and install and use
> it. If you can demonstrate a problem there, now people will be much more
> interested in addressing the issue.
> 
> Mark
> 
>>
>>  Regards,
>>  Lipi
>>
>>  Lipi Thukral wrote:
>>  > Dear gmx-users,
>>  >
>>  > I intended to concantanate two trajectories with no similar time frames
>>  > i.e., both files have distinct time frames. But once I combine the files
>>  > many time frames are skipped, details are as follows:
>>  >
>>  > File 1 has 2175 frames
>>  > File 2 has 1232 frames
>>  >
>>
>>  What does gmxcheck tell you about each of these files?
>>
>>  > I used the following command to concatenate:
>>  >
>>  > trjcat -f File1.xtc File2.xtc -o CombinedFile3.xtc
>>  >
>>  > Now I obtained in total 1266 frames instead of 3407 frames as expected.
>>  > When I compared the CombinedFile3 with both File 1 and File 2, I found
>>  > that it has somehow taken the first  34 frames from File 1 and rest of
>>  > the frames belongs to File 2 i.e., the last file.
>>  >
>>  > I am unable to understand why trjcat skips the frames inspite of no
>>  > similarity. I am using 4.5-beta version of Gromacs. I am also open to
>>  > sharing the files, should that be needed.
>>  >
>>
>>  There have been numerous beta versions over the last few years, none
>>  of which
>>  should necessarily be considered stable.  Which beta are you using
>>  exactly?
>>
>>  -Justin
>>
>>  > Any possible explanation as to why I am obtaining the combined
>>  > trajectory with skipping frames, would be really helpful.
>>  >
>>  > Many Thanks,
>>  > Lipi
>>  >
>>
>>  --
>>  ========================================
>>
>>  Justin A. Lemkul
>>  Ph.D. Candidate
>>  ICTAS Doctoral Scholar
>>  MILES-IGERT Trainee
>>  Department of Biochemistry
>>  Virginia Tech
>>  Blacksburg, VA
>>  jalemkul[at]vt.edu <http://vt.edu/> <http://vt.edu/> | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>>  ========================================
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================