Re: [AMBER] cpptraj hist 3D

From: andrea spitaleri <andrea.spitaleri.iit.it>
Date: Thu, 06 Nov 2014 15:38:23 +0100

In addition, you may try python with matplotlib/scipy:

http://docs.scipy.org/doc/numpy/reference/generated/numpy.histogramdd.html

hope it helps

and



On 06/11/2014 15:29, Jason Swails wrote:
>> On Nov 6, 2014, at 7:54 AM, CHAMI F. <fatima.chami.durham.ac.uk> wrote:
>>
>> Dear Amber user,
>>
>> is there a way to represent a 3D histogram from output of hist in cpptraj
>>
>> hist phi1 phi2 phi3 min -180 max 180 bins 72 out hist_id_20.gnu
>>
>>
>> 2D histogram is fine in gnuplot using: splot "-" with pm3d title “hist.gnu"
> N-dimensional data requires N+1 dimensions in order to display the full function. That is why a 2D histogram appears like a surface.
>
> By extension, a 3D histogram requires 4 dimensions to plot the full function. In a standard plot, this is clearly impossible. The trick then becomes how to reduce the dimensionality to 3 or fewer dimensions so you can plot it.
>
> A common approach is to simply plot a relevant contour map, so plot the surface representing a particular value of the histogram, as a function of the three dihedrals (this is what is done, for instance, with wavefunctions and orbitals in electronic structure calculations). You can plot multiple values at different transparencies, similar to a contour map in 2 dimensions of, say, a mountain range.
>
> Another approach is to plot “slices” -- that is, plot a surface of phi1 and phi2 for incremental values of phi3. The best approach is the one that most clearly conveys the conclusions you wish to convey from the data.
>
> I’ve never tried to do this myself, and the only software I know of that does this “easily” is Mathematica (although Octave and Matlab documentation suggests it’s possible there). I don’t know if gnuplot can do this.
>
> HTH,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber

-- 
Andrea Spitaleri PhD
D3 - Drug Discovery & Development
Istituto Italiano di Tecnologia
Via Morego, 30 16163 Genova
cell: +39 3485188790
http://www.iit.it/en/d3-people/andrea-spitaleri.html
ORCID: http://orcid.org/0000-0003-3012-3557
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Nov 06 2014 - 07:00:02 PST
Custom Search