On Tue, Jan 31, 2017 at 9:16 AM, Vlad Cojocaru
<vlad.cojocaru.mpi-muenster.mpg.de> wrote:
>
> If I understand correctly, this means that CHARMM changed the DCD
> formatting which caused a recent update of CPPTRAJ to read those. But
> this update made CPPTRAJ unable to correctly process NAMD DCDs because
> NAMD did not update to the new CHARMM formatting.
The CHARMM format has actually stored the shape matrix instead of unit
cell parameters since version 22. When I wrote the cpptraj Charmm DCD
trajectory parser I erroneously omitted the conversion to and from the
shape matrix (note that I think ptraj may have handled this correctly
so this is my mistake). I'm not sure what the history of Charmm DCD
files is with NAMD, so I'm not sure if it was a conscious decision on
their part to just store unit cell parameters. If it was then they
should decrement the Charmm version number in the trajectory to at
least 21 to reflect this (which is what cpptraj now does when writing
unit cell parameters to Charmm trajectory files).
>
> Am I right here ? If yes, do you know which is precisely the update that
> has changed CPPTRAJ behavior towards DCDs ? As I mentioned before, a
> CPPTRAJ version compiled in July 2016 reads and processes NAMD DCDs
> absolutely fine.
There are two related pull requests:
https://github.com/Amber-MD/cpptraj/pull/399
https://github.com/Amber-MD/cpptraj/pull/427
Looks like the former was incorporated as update 16 while the latter
hasn't yet made it to AmberTools.
-Dan
>
> Best wishes
> Vlad
>
>
>
>
> On 01/31/2017 02:07 PM, Daniel Roe wrote:
>> On Tue, Jan 31, 2017 at 7:32 AM, Vlad Cojocaru
>> <vlad.cojocaru.mpi-muenster.mpg.de> wrote:
>>> Actually adding the "ucell" keyword to the "trajin" command in the
>>> github version of cpptraj solves the issue (but only in the github
>>> version) ....
>> Glad to hear the GitHub version works for you.
>>
>>> Is this update planned to be included in Amber16 soon ?
>> The convention for cpptraj GitHub -> AmberTools has been to only port
>> over bug fixes to cpptraj and leave new functionality for AmberTools
>> releases (next one should be around April or so I think). Technically
>> this is a case of NAMD not following the CHARMM format and not a bug
>> in cpptraj :-), but if I get a chance I'll try to make a fix (sadly I
>> don't have as much time as I'd like for cpptraj development anymore).
>> Until then hopefully the GitHub version of cpptraj will work for you.
>>
>> Thanks for the report,
>>
>> -Dan
>>
>>> Best
>>> Vlad
>>>
>>> On 01/31/2017 01:16 PM, Vlad Cojocaru wrote:
>>>> Hi Dan,
>>>>
>>>> Thanks for your feed-back ...
>>>>
>>>> The trajectories were calculated with NAMD 2.11 and 2.12 ... I thought
>>>> of doing one last check using the github version of cpptraj with the
>>>> "ucell" keyword ... If this also does not solve the issue, I will
>>>> provide you the top and trajectory file off the list
>>>>
>>>> Best
>>>> Vlad
>>>>
>>>>
>>>> On 01/31/2017 12:54 PM, Daniel Roe wrote:
>>>>> Hi,
>>>>>
>>>>> Debugging this is going to be tricky without access to the trajectory or
>>>>> one like it. I assume this trajectory was not generated with charmm
>>>>> (if it
>>>>> was let me know). It sounds like the trajectory may not correspond
>>>>> exactly
>>>>> to the charmm format (which to be fair is not exactly well documented).
>>>>> Contact me off list if there is some way I can get access to this or
>>>>> maybe
>>>>> send me a smaller trajectory/topology that exhibits the same behavior.
>>>>> Thanks,
>>>>>
>>>>> -Dan
>>>>>
>>>>> On Tue, Jan 31, 2017 at 4:10 AM Vlad Cojocaru <
>>>>> vlad.cojocaru.mpi-muenster.mpg.de> wrote:
>>>>>
>>>>>> Dear all,
>>>>>>
>>>>>> I am having an issue with cpptraj when trying to image DCD trajectories
>>>>>> back into a truncated octahedron box. I am uising Amber 16 on a
>>>>>> supercomputer and as its a brand new compilation, all updates are
>>>>>> installed. I have an amber topology (parm7) and a number of DCD
>>>>>> trajectories. "autoimage familiar" actually gives me a strange cube
>>>>>> instead of the normal truncated octahedron.
>>>>>>
>>>>>> Now when I run the same job on my workstation with a version of
>>>>>> Amber 16
>>>>>> compiled in July 2016, I actually get the correct truncated octahedron.
>>>>>> Therefore, I suspect this might have something to do with some very
>>>>>> recent updates of cpptraj and the way the program reads the information
>>>>>> from DCD files.
>>>>>>
>>>>>> In the cpptraj output I found these lines which may be relevant
>>>>>>
>>>>>> Warning: oct4_sox2_hoxb1_mur_01_box_cmd0101.dcd: Reported number of
>>>>>> frames in DCD file is 0,
>>>>>> Warning: actual number of frames is 6000. Only reading 6000
>>>>>> frames.
>>>>>> Version >= 22; assuming shape matrix is stored.
>>>>>> Warning: Trajectory box type is 'Non-orthogonal' but topology box type
>>>>>> is 'Trunc. Oct.'.
>>>>>> Warning: Setting topology box information from trajectory.
>>>>>>
>>>>>> I also found an earlier related thread in which Dan Roe suggested to
>>>>>> use
>>>>>> the "ucell" keyword for the "trajin" command as a fix but this does not
>>>>>> fix the issue.
>>>>>>
>>>>>> I'd appreciate knowing if there is a way to fix this behavior or if I
>>>>>> need to revert some of the updates.
>>>>>>
>>>>>> Thanks a lot for answering
>>>>>>
>>>>>> Best wishes
>>>>>> Vlad
>>>>>>
>>>>>> --
>>>>>> Vlad Cojocaru, Ph.D., Project Group Leader
>>>>>> Department of Cell and Developmental Biology
>>>>>> Max Planck Institute for Molecular Biomedicine
>>>>>> Röntgenstrasse 20, 48149 Münster, Germany
>>>>>> Tel: +49-251-70365-324; Fax: +49-251-70365-399
>>>>>> Email: vlad.cojocaru[at]mpi-muenster.mpg.de
>>>>>> http://www.mpi-muenster.mpg.de/43241/cojocaru
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> AMBER mailing list
>>>>>> AMBER.ambermd.org
>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>>
>>> --
>>> Vlad Cojocaru, Ph.D., Project Group Leader
>>> Department of Cell and Developmental Biology
>>> Max Planck Institute for Molecular Biomedicine
>>> Röntgenstrasse 20, 48149 Münster, Germany
>>> Tel: +49-251-70365-324; Fax: +49-251-70365-399
>>> Email: vlad.cojocaru[at]mpi-muenster.mpg.de
>>> http://www.mpi-muenster.mpg.de/43241/cojocaru
>>>
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>
> --
> Vlad Cojocaru, Ph.D., Project Group Leader
> Department of Cell and Developmental Biology
> Max Planck Institute for Molecular Biomedicine
> Röntgenstrasse 20, 48149 Münster, Germany
> Tel: +49-251-70365-324; Fax: +49-251-70365-399
> Email: vlad.cojocaru[at]mpi-muenster.mpg.de
> http://www.mpi-muenster.mpg.de/43241/cojocaru
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
-------------------------
Daniel R. Roe
Laboratory of Computational Biology
National Institutes of Health, NHLBI
5635 Fishers Ln, Rm T900
Rockville MD, 20852
https://www.lobos.nih.gov/lcb
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jan 31 2017 - 10:00:04 PST
