A tool to boost protein functional dynamics with colored contact maps
HOPMA is intended to help to define elastic network representations of protein structures. It takes as input a protein 3D structure and gives as output a set of protein region pairs that should not be connected by any link in the network. HOPMA exploits the protein inter-residue distance matrix. It transforms it into a smoothed binarized colored contact map and detects small contiguous patches in this map. These patches correspond to small contact areas between contiguous protein segments, and the rational is that they indicate contacts that lock the elastic network and hinder its motions. HOPMA's output can be directly given to NOLB: https://team.inria.fr/nano-d/software/nolb-normal-modes/.
The fastest way to install HOPMA is to download its Singularity image hopma.sif.
To use the tool, simply type:
./hopma.sif 1dap B
where 1dap is the input PDB filename without the extension and B is the name of the chain you want to analyze.
You can find the corresponding PDB file 1dap.pdb in the example directory of this repository.
You can clone this HOPMA package using git:
git clone https://github.com/elolaine/HOPMA.git
HOPMA is implemented in Python and R. https://www.python.org/ https://cran.r-project.org/
To be able to use the tool, you need to install the BioPython and Numpy Python modules.
- for BioPython: pip install Bio
- for R: brew install r OR brew install --cask r
Define and export the environment variable HOPMA_PATH=/path-to-HOPMA-directory/.
Test the program on the example PDB file 1dap.pdb, by typing:
python3 $HOPMA_PATH/hopma.py 1dap -c B
The outputs should be identical to those stored in the example directory.
A help can be accessed by typing:
python $HOPMA_PATH/hopma.py --help
The protein 3D structure filename in PDB format without the .pdb extension is a mandatory argument.
You may also give the name of the protein chain you want to analyse ('A' by default).
By default, HOPMA will compute a smoothed contact propensity map using windows of size 5x5 (padding = 2). It will then use two different contact propensities cutoff values, 0.1 and 0.001, to generate two colored binarized maps. The output list of excluded contacts is determined by making the union of the small patches (size lower than 625 cells) detected in the two binarized maps.
HOPMA's outputs are:
- the CA-CA distance matrix (.dist) in text format,
- Two colored binarized contact maps (.class) in text format,
- Two PDF files showing images of the colored binarized contact maps,
- Three text files (.excl) containing lists of pairs of protein regions that should not be linked by any spring.
The final result is the "combi" list. This file can be given to NOLB with the option --excl.
Laine E, Grudinin S. (2021) HOPMA: Boosting Protein Functional Dynamics with Colored Contact Maps. J Phys Chem B.