Ramachandran

Ramachandran plot visualizes energetically allowed regions for backbone torsion (dihedral) angles ψ against φ of amino acid residues in protein structure. This Python package provides the implementations from scratch to crawling PDB/PDBx files from RCSB PDB Bank, parsing PDB/PDBx files for extracting geometric information, computing the torsion angles of residues from atom coordinates, analyzing the Ramachandran favorable regions from the reference PDB/PDBx files selected, and creating Ramachandran plots for user-provided PDB/PDBx files.

Dependencies

Python 3.8+
Numpy 1.20.1+
Scipy 1.6.0+
Matplotlib 3.3.4+

Usages

It is recommended to use Docker container to run the program.

Build Docker Image

$ docker build -f docker/ramachandran.Dockerfile --no-cache --tag=ramachandran:0.0.1 .

Run Docker Container

$ docker run -it --rm --gpus device=0 -v $(pwd):/mnt ramachandran:0.0.1

Install Package

$ pip install ramachandran

Create Ramachandran Plots

$ ramachandran-plot --help
usage: ramachandran-plot [-h] --file-path FILE_PATH
                         [--save-dir-path SAVE_DIR_PATH]
                         [--protein-name PROTEIN_NAME]

Ramachandran Plot Tool. Plots for GLY, PRO, pre-PRO, and Other residues will
be created.

optional arguments:
  -h, --help            show this help message and exit
  --file-path FILE_PATH
                        PDB/PDBx file path. (default: None)
  --save-dir-path SAVE_DIR_PATH
                        Directory path for saving the Ramachandran plots for
                        the PDB/PDBx file. (default: .)
  --protein-name PROTEIN_NAME
                        Protein name. (default: None)

Demo

$ ramachandran_plot --file-path 1AXC.cif --save-dir-path 1AXC --protein-name 1AXC

General	GLY	PRO	Pre-PRO

TODO

Filter homologous proteins for computing the Ramachandran statistics, since too many homologous protein structures will bias the statistics.
Add the original hard-coded Ramachandran favorable region.

leimao/Ramachandran