Hi Daniel
Sorry but I believe intrdf has some bug ( or please ignore if it was kept
intentionally to make user aware what he/she is doing)
As you must have the data with you (for the rdf you plotted):
radial rdf_all.dat 0.1 10 :WAT (:484,441,382,213>.100&.Cl-) intrdf
int_all.dat
In this case RDF is averaged nicely but intrdf is not (you can also see
water goes upto 10e6, though I know it is integral but I would expect to
give an average). As I checked, simple fix is to divide all values by
#mask2 (numbers in mask2).
vs
radial rdf_1.dat 0.1 10 :WAT :#1Cl- intrdf int1.dat
radial rdf_2.dat 0.1 10 :WAT :#2Cl- intrdf int2.dat
& so on, finally average all int*.dat and rdf_*.dat
Here in this case rdf_*.dat when averaged should be same as rdf_all.dat
(which you also got same) but the average of int*dat is not same as
int_all.dat (but one has to divide by #mask2 to get same values as in
averaged rdf_*dat)
kind regards
jiomm
On Fri, Apr 11, 2014 at 8:57 PM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
> On Fri, Apr 11, 2014 at 1:13 PM, newamber list <newamberlist.gmail.com>
> wrote:
> > Finally one thing, as now I am using a nonimaged traj with 'reference' ,
> so
> > 'radial' will consider the imaged or nonimaged referenced frame?
>
> The radial command will image distances whenever periodic boundary
> conditions (i.e. box information) are present in the trajectory unless
> 'noimage' is specified. The 'reference' keyword in no way affects
> 'radial'.
>
> -Dan
>
> >
> > reference test.nc 10
> > trajin test.nc 10 10
> > radial rdf.dat 0.1 10 :WAT (:484,441,382,213>.100&.Cl-) volume
> >
> > thanks
> > JIomm
> >
> >
> > On Fri, Apr 11, 2014 at 7:33 PM, Daniel Roe <daniel.r.roe.gmail.com>
> wrote:
> >
> >> Hi,
> >>
> >> I can't reproduce the behavior you are seeing. When I generate similar
> >> plots, I get a nice overlap between single RDFs and averaged RDF (see
> >> attached). I don't think this would be an averaging bug anyway - the
> >> fact that you get population at the lower distance bins means the
> >> underlying histogram had some data there. If you are absolutely
> >> certain none of those distances should be shorter than 2 Ang., the
> >> only thing I can think of is maybe you are using a trajectory with
> >> periodic boundary conditions that has been rotated e.g. via an RMS
> >> fit, which would cause imaging artifacts. What happens if you specify
> >> 'noimage' to the 'radial' command?
> >>
> >> -Dan
> >>
> >>
> >> On Thu, Apr 10, 2014 at 4:08 PM, newamber list <newamberlist.gmail.com>
> >> wrote:
> >> > Hi All
> >> >
> >> > I am getting some unusual rdf starting at zero distances of ion which
> is
> >> > wrong and I think could be an averaging problem. Here is a single
> frame
> >> > command. (and plot attached)
> >> >
> >> > reference test.nc 10
> >> >
> >> > trajin test.nc 10 10
> >> >
> >> > radial rdf.dat 0.1 10 :WAT (:484,441,382,213>.100&.Cl-) volume
> >> >
> >> >
> >> > Actually this should give me an average RDF for all the Cl- ions
> which
> >> are
> >> > 100 Ang. distance far from COM of :484,441,382,213. There should be no
> >> RDF
> >> > values below distance 2 or so. But when I test the same with some
> >> > individual Cl- ions (>100 Angs) with masks like this for individual
> >> ion:WAT
> >> > :3740, I don't get any such RDF near distance zero which is correct.
> >> >
> >> >
> >> > So I think there is some averaging bug?
> >> >
> >> >
> >> > Thanks
> >> >
> >> > Jiomm
> >> >
> >> > _______________________________________________
> >> > 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
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 11 2014 - 17:00:02 PDT
