Hey,
thanks Dan for your reply, sorry to bother u guys again.
I rechecked this stuff, doing a simple experiment in ptraj:
reference ref.pdb
trajin prorun_reimaged.mdcrd 1 1
strip !.CA,C,N
rms reference .CA out test.list
trajout prorun_reimaged_mod.mdcrd pdb
I load in my trajectory, strip everything except the backbone trace and
then do the rms, where I still argue he does translate.
When I look at the pdb: prorun_reimaged_mod.mdcrd.pdb I see a structure
that is nearer to the ref.pdb then to the original frame from the
trajectory,
albeit not really fitted, more placed in the area. So he obvioulsy
fitted something, but not doing it correctly. But when I use my inpcrd
superimpose it in pymol onto my wanted reference save it and then use
this structure as reference, he does a good job. So I guess the reason
is that my reference has no H and no Wat and the trajectory has ....
despite I use only the CA trace for superimposition. Any ideas?
On 05/09/2013 09:41 PM, Daniel Roe wrote:
> Hi,
>
> One of the strengths of cpptraj is that it allows you to specify
> multiple topologies, as well as different target and reference masks
> to the rms command - the only requirement is that the # of atoms and
> their ordering are the same. For example, the input may be something
> like (assuming :63-65.CA,C,N is your target mask and
> :94,146,220.CA,C,N is your reference mask):
>
> parm ref_structure.pdb [p_ref] # This loads ref_structure.pdb as a
> topology tagged [p_ref]
> reference ref_structure.pdb parm [p_ref] # This loads the coordinates
> from ref_structure.pdb as the reference
> trajin prorun_reimaged.mdcrd
> rms reference :63-65.CA,C,N :94,146,220.CA,C,N out results.dat
>
> On Thu, May 9, 2013 at 11:02 AM, Andre C. Stiel
> <andre.stiel.tuebingen.mpg.de> wrote:
>> 2) the rmsd results are absolutely bogus. It looks like in the first
>> step where I did NOT specified "nofit" he
>> did no fitting or he didnt translate the frame.... I always figured
>> ptraj would translate the frame if no "nofit" is specified
> No, in ptraj and cpptraj 'nofit' means straight RMSD, do not modify
> the coordinates in any way. In cpptraj you can get RMS results with
> translation and without rotation by specifying the 'norotate' keyword.
>
> Hope this helps,
>
> -Dan
>
>> In short: I want to first superimpose a set of relatively fixed atoms
>> and then measure the rmsd w/o fitting of my moving atoms.
>>
>> Maybe somebody can help me out.
>> Thanks a lot
>>
>> André
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri May 10 2013 - 05:30:02 PDT
