Overlapy

Project for the subjects of Structural Bioinformatics and Python (Master in Bioinformatics for Health Sciences Universitat Pompeu Fabra - Universitat de Barcelona); developed by Álvaro Abella Bascarán, Josep Arús Pous and Samuel Miravet Verde.

What is Overlapy?

Overlapy is a python library and command line tool which provides a easy way to do protein superimposition. Overlapy is also provided in an even more user-friendly way, as a web application.

If you are interested in know more about the program, its implementation and several examples of use, please read our report (you have to download it as raw data in format pdb).

Installation

Overlapy uses python3. Please check that it is installed:

python3 --version

If it's not, install it.

sudo apt-get install python3.4

In order to get Overlapy to work, you must first satisfy some library requirements. The preferred way is to use pip for this:

sudo apt-get install python3-pip
pip3 install -r requirements.txt

Once this is done, you can install the library + command line tool:

sudo python3 setup.py install

Now overlapy is available for you! You can use it as a python3 module, or as a command line tool:

Overlapy as a python module. Usage example

import overlapy
# instantiate the Superimposer class
superimposer = overlapy.Superimposer()
# feed it the desired pdb files
superimposer.parse("file1.pdb", "file2.pdb")
# let's superimpose chains H and L with chains H and L
superimposer.select_chains(["H", "L"], ["H", "L"])
superimposer.superimpose()
# now let's get the results
rmsd = superimposer.rmsd
rotation_matrix = superimposer.rotation_matrix
alignment = superimposer.get_multiple_sequence_alignment()
# we can save the superimposed structure:
superimposer.save_superimposed_pdb("outputfile.pdb")

Overlapy as a command line tool.

Once you have run setup.py, you can execute it as a command line tool:

overlapy --help

To perform a basic superimposition just run:

overlapy -i1 protein1.pdb -i2 protein2.pdb -o protein1protein2.pdb

This will generate a PDB file containing the two initial structures superimposed (stored the input 1 as chain A and chain B for the second) and will show in your terminal the best Needleman & Wunsch alignment, the rotation matrix and the RMSD.

If you want to store in files the matrix (in tsv), the alignment (in clustal) and/or the RMSD (text file), just activate the options -r, -m and -a.

Finally, if you want to select chains to superimpose from the input PDBs, use the -c1 and -c2 arguments:

overlapy -i1 protein1.pdb -c1 HL -i2 protein2.pdb -c2 A -o protein1protein2.pdb -a -m