Ramaplot is a Python program used to generate Ramachandran plot <https://en.wikipedia.org/wiki/Ramachandran_plot> figures using matplotlib.
Ramachandran plots are a powerful method of visualizing the distribution of
backbone Φ and Ψ dihedral angles in peptides and proteins, or of other
properites as a function of Φ and Ψ. Ramaplot configuration is powered by
myplotspec <http://karltdebiec.github.io/myplotspec>, enabling powerful
configuration options from YAML <http://www.yaml.org/spec/1.2/spec.html>_-format text files. Several types of
experimental and simulated datasets are currently supported.
.. image:: examples/example.png
Experiment
Neighbor-Dependent Ramachandran Distributions
_____________________________________________
The Neighbor-Dependent Ramachandran Distribution (NDRD) dataset is a collection
of Φ/Ψ probability distributions measured from the loop (non-α/β) regions of
solved structures. In addition to distributions for all 20 amino acids, left-
and right-neighbor distributions are available for each pair of amino acids,
which may be used to calculate distributions of each amino-acid triplet. The
complete dataset includes residues identified as turn, coil, bridge, π helix,
and 3\ :sub:`10` helix; versions including only turn, coil, and bridge, only turn, and
only coil are available as well. This dataset may be obtained from
`<http://dunbrack.fccc.edu/ndrd>`_.
`Ting, Daniel, Wang, Guoli, Shapovalov, Maxim, Mitra, Rajib, Jordan, Michael I,
Dunbrack Jr. Roland L. Neighbor-Dependent Ramachandran Probability
Distributions of Amino Acids Developed from a Hierarchical Dirichlet Process
Model. PLoS Computational Biology. 2010. 6. e1000763.
<http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1000763>`_
Conformation-Dependent Library
______________________________
The Conformation-Dependent Library (CDL) dataset is a collection of averages
and standard deviations of backbone bond lengths, angles, and ω as a function
of Φ and Ψ, measured from solved structures. The dataset is divided into 8
classes: isoleucine/valine, glycine, proline, and all other residues, each
leading into proline or residues other than proline. This dataset may be
obtained from `<http://dunbrack.fccc.edu/nmhrcm>`_ and
`<http://dunbrack.fccc.edu/omega>`_.
`Berkholz, Donald S., Shapovalov, Maxim V., Dunbrack Jr. Roland L., Karplus, P.
Andrew. Conformation Dependence of Backbone Geometry in Proteins. Structure.
2009. 17. 1316-1325.
<http://www.sciencedirect.com/science/article/pii/S0969212609003359>`_
`Berkholz, Donald S., Driggers, Camden M., Shapovalov, Maxim V., Dunbrack Jr.,
Roland L., Karplus, P. Andrew. Nonplanar Peptide Bonds in Proteins are Common
and Conserved but not Biased Toward Active Sites. Proceedings of the National
Acadedemy of the Sciences of the United States of America. 2012. 109. 449-453.
<http://www.pnas.org/content/109/2/449>`_
SimulationMolecular Dynamics Simulation
The distribution of observed Φ and Ψ dihedral angles from a molecular dynamics
simulation may be calculated from a text file containing a timeseries,
typically calculated from the trajectory using cpptraj <https://github.com/Amber-MD/cpptraj>_. Alternatively, the average value of
another property as a function of Φ and Ψ may be calculated.
Roe, Daniel R., Cheatham III, Thomas E. PTRAJ and CPPTRAJ: Software for Processing and Analysis of Molecular Dynamics Trajectory Data. Journal of Chemical Theory and Computation. 2013. 93084–3095. <http://pubs.acs.org/doi/abs/10.1021/ct400341p>_
Umbrella Sampling/Weighted Histogram Analysis Method
As an alternative to brute-force sampling of a simulated protein system using a
long continuous unbiased trajectory, it is possible to obtain comparable
results in less simulation time using the Umbrella Sampling method of enhanced
sampling. In this method, the system is restrained to selected values along one
or more coordinates of interest (such as the Φ and Ψ dihedral angles),
typically using a harmonic restraint. Provided that the sampling within each
'window' overlaps with its neighbors, the true free energy of the system as a
function of the chosen coordinate(s) may be subsequently calculated using the
Weighted Histogram Analysis Method (WHAM) <http://membrane.urmc.rochester.edu/content/wham>_
Grossfield, Alan. WHAM: The Weighted Histogram Analysis Method. Version 2.0.9. <http://membrane.urmc.rochester.edu/content/wham>_
Molecular Mechanics Force Fields
The distribution of Φ and Ψ dihedral angles sampled in a molecular dynamics
simulation are a function of the force field parameters used, in particular the
fourier series used to model the potential energy surface along Φ and Ψ. These
parameters may be loaded from an Amber-format force field <http://ambermd.org/#ff>_ (parm*.dat).
(Work in progress)
Images
In order to enable more direct comparison to published Ramachandran plots,
Ramaplot supports plotting of image files.
Example Usage
-------------
The example output shown above may be generated using the following command:
::
python Ramaplot.py -yaml examples/example.yml
example.yml::
preset: notebook
figures:
0:
outfile: examples/example.png
ncols: 3
bottom: 0.9
subplots:
0:
preset: colorbar_bottom
title: AMBER ff14SB
datasets:
0:
kind: wham
infile: examples/ff14SB.dat
1:
title: CHARMM36
datasets:
0:
kind: wham
infile: examples/charmm36.dat
2:
preset: [diff, colorbar_bottom]
title: ff14SB - CHARMM36
datasets:
0:
dataset_kw:
minuend:
kind: wham
infile: examples/ff14SB.dat
subtrahend:
kind: wham
infile: examples/charmm36.dat
mask_cutoff: 5Ramaplot supports Python 2.7 and 3.4, and requires the following packages:
Ramaplot has been tested with Anaconda Python 2.2.0 on Arch Linux, OSX Yosemite, and Windows 8.1.
Ramaplot is developed by Karl T. Debiec, a graduate student at the University of Pittsburgh advised by Professors Lillian T. Chong and Angela M. Gronenborn.
Released under a 3-clause BSD license.