Hi,
On Sat, Jun 21, 2014 at 7:34 PM, newamber list <newamberlist.gmail.com>
wrote:
> Sorry looks like I messed up everything. Actually I want to do similar
> analysis done with gromacs '' filter the trajectory to show only the motion
> along eigenvectors"
>
If I understand the purpose of this functionality correctly, this is not
currently implemented in cpptraj. You can only obtain the projection of
coordinates along specified eigenvectors (the 'projection' action).
> readdata evecs.dat
> crdaction crd1 projection modes evecs.dat out myproj.txt beg 1 end
> 1 :1-382.N
> readdata myproj.txt
> filter myproj.txt min -10 max 10 out filter.dat
> trajout filter.nc
>
> So if it is to be motion along eigenvector 1 then I should choose max and
> min values (Mode1) from myproj.txt ?
>
First, you don't need to read myproj.txt back in - in fact during
processing it will not have been generated yet so this statement should be
removed. You can then give the projection data set a name in the
'projection' action (e.g. "myprojection") and refer to it that way, like so:
readdata evecs.dat
crdaction crd1 projection myprojection modes evecs.dat out myproj.txt beg 1
end 1 :1-382.N
filter myprojection:1 min -10 max 10 out filter.dat
trajout filter.ev1.nc
This will give you all frames with projection values between -10 and 10 for
the first eigenvector. Note that I am using a data set index (":X") in the
'filter' command; in this case since you are only generating one projection
it is unnecessary, but if you have more than one projection this will
select the one you want (1 for projection along eigenvector 1, 2 for
eigenvector 2, etc).
Note that the frames will contain contributions from other modes as well.
You could try to filter some of them out with additional filter commands,
like so:
crdaction crd1 projection myprojection modes evecs.dat out myproj.txt beg 1
end 10 :1-382.N
filter myprojection:1 min -10 max 10 out filter.dat
filter myprojection:2 min -1 max 1 out filter.dat
filter myprojection:3 min -1 max 1 out filter.dat
etc. However, you may not have a lot of frames that fit this criteria so it
may not work so well.
Also one more thing does pseudo-trajectory (from 'modes trajout') means an
> interpolation between two extremes?
>
What the pseudo-trajectory option does is take the average structure
(generated during the matrix creation step and stored in your evecs output
file) and project it along a specified mode with respect to the minimum and
maximum projection values you specify. So in that sense it is an
interpolation between specified extremes.
Hope this helps,
-Dan
>
>
> thanks
>
>
> On Sat, Jun 21, 2014 at 9:48 PM, Daniel Roe <daniel.r.roe.gmail.com>
> wrote:
>
> > Hi,
> >
> > On Sat, Jun 21, 2014 at 11:07 AM, newamber list <newamberlist.gmail.com>
> > wrote:
> >
> > > crdaction crd1 projection modes evecs.dat out myproj.txt beg 1 end 8
> > > :1-382.N
> > > modes trajout filter_1.nc trajoutfmt netcdf name test pcmin -100 pcmax
> > 100
> > > tmode 1
> > >
> > > I have following problems/queries:
> > >
> > > 1) I always get 201 frames. I am not sure why is it so? No matter how
> > many
> > > frames are there in trajin and which mode I trajout, its always 201.
> > >
> >
> > Note that for the 'modes' analysis, 'trajout' is a keyword requesting
> > generation of a pseudo-trajectory along a specified eigenvector. In this
> > case you are requesting that the 'modes' analysis create a psuedo
> > trajectory along your first eigenvector ranging from a projection value
> of
> > -100 to 100; (100 - -100 + 1) = 201 frames. You should also ensure that
> the
> > minimum and maximum projection values are reasonable (by e.g. creating a
> > histogram of the projection values from your input trajectories).
> >
> >
> > >
> > > 2) Also how I can get frame number from input trajectory (trajin) which
> > are
> > > filtered out and saved in some output trajectory (modes trajout)
> > >
> >
> > It's not clear from your description what exactly you are trying to
> filter.
> > Do you want frames that fall within certain principal component
> projection
> > values? If so you need to use the 'filter' action on the data set(s) you
> > get from the 'projection' action, followed by a 'trajout' command.
> >
> >
> > >
> > > 3) Also please let me know if I understood it correctly: If am not
> wrong
> > > then the filtered trajectory so obtained should contain the extreme
> > > projections? Thus the two extreme projections should 'reflect' the
> filter
> > > trajectory in summary.
> > >
> >
> > I'm not sure I understand this question completely, but the first and
> last
> > frames of your pseudo-trajectory (from 'modes trajout ...') will
> correspond
> > to projection values <pcmin> and <pcmax> along the specified eigenvector.
> >
> > Hope this helps,
> >
> > -Dan
> >
> >
> > >
> > > Thanks for any help
> > >
> > > regards
> > > JIom
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> >
> >
> >
> > --
> > -------------------------
> > Daniel R. Roe, PhD
> > Department of Medicinal Chemistry
> > University of Utah
> > 30 South 2000 East, Room 201
> > Salt Lake City, UT 84112-5820
> > http://home.chpc.utah.edu/~cheatham/
> > (801) 587-9652
> > (801) 585-6208 (Fax)
> > _______________________________________________
> > 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
>
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 24 2014 - 09:00:03 PDT
