DisVis

About DisVis

DisVis is a Python package and command line tool to visualize and quantify the accessible interaction space of distance restrained binary biomolecular complexes. It performs a full and systematic 6 dimensional search of the three translational and rotational degrees of freedom to determine the number of complexes consistent with the restraints. In addition, it outputs the percentage of restraints being violated and a density that represents the center-of-mass position of the scanning chain corresponding to the highest number of consistent restraints at every position in space.

Requirements

• Python 3.11+
• A C compiler (to build the extension modules)

The Python packages DisVis needs are listed in pyproject.toml and installed automatically by pip. GPU acceleration also needs an OpenCL 1.2+ enabled GPU and the optional gpu packages (see INSTALLATION.md).

Installation

DisVis is best installed in a virtual environment. Open a terminal and type

python3 -m venv .venv
source .venv/bin/activate
pip install disvis

After installation, the command line tool disvis should be at your disposal. To install from the source instead, clone the repository and run pip install . in it.

For GPU acceleration, install the optional GPU requirements with

pip install 'disvis[gpu]'

See INSTALLATION.md for the requirements and for checking the GPU set-up.

Linux

Make sure git, the Python header files and a C compiler are available, e.g. on Debian/Ubuntu

sudo apt-get install git python3-dev python3-venv build-essential

If you are working on a Fedora system, use dnf and the python3-devel and gcc packages. Follow the general instructions above to install DisVis.

MacOSX

Install the Xcode command line tools, which provide git and a C compiler

xcode-select --install

and a recent Python 3, e.g. with brew

brew install python

Follow the general instructions above to install DisVis.

Windows

Install git for Windows, a Python 3 distribution, such as Anaconda, and the Microsoft C++ Build Tools. Follow the general instructions above to install DisVis.

Usage

The general pattern to invoke disvis is

disvis <pdb1> <pdb2> <distance-restraints-file>

where <pdb1> is the fixed chain, <pdb2> is the scanning chain and <distance-restraints-file> is a text-file containing the distance restraints in the following format

<chainid 1> <resid 1> <atomname 1> <chainid 2> <resid 2> <atomname 2> <mindis> <maxdis>

As an example

A 18 CA F 27 CB 10.0 20.0

This puts a distance restraint between the CA-atom of residue 18 of chain A of pdb1 and the CB-atom of residue 27 of chain F of pdb2 that should be longer than or equal to 10A and smaller than or equal to 20A. Comments can be added by starting the line with the pound sign (#) and empty lines are ignored.

Options

To get a help screen with available options and explanation type

disvis --help

Some examples to get you started. To perform a 5.27 degree rotational search and store the results in the directory results/

disvis 1wcm_A.pdb 1wcm_E.pdb restraints.dat -a 5.27 -d results

Note that the directory is created if it doesn't exist.

To perform a 9.72 degree rotational search with 16 processors and a voxel spacing of 2A

disvis O14250.pdb Q9UT97.pdb restraints.dat -a 9.72 -p 16 -vs 2

To offload computations to the GPU (if the requirements are met) and increase the maximum allowed volume of clashes as well as the minimum required volume of interaction, and set the interaction radius to 2A

disvis 1wcm_A.pdb 1wcm_E.pdb restraints.dat -g -cv 6000 -iv 7000 -ir 2

To perform a grid occupancy analysis for complexes consistent with at least N restraints, include the -oa or --occupancy-analysis flag to the command. By default, the analysis is only performed on complexes consistent with either all and all - 1 restraints. If you want to perform an occupancy analysis for complexes consistent with a lower number of restraints, use the -ic or --interaction-restraints-cutoff option, followed by the minimum number of required consistent restraints. Thus the command

disvis 1wcm_A.pdb 1wcm_E.pdb restraints.dat -oa -ic 5

will make an occupancy grid for complexes consistent with at least 5 restraints, and higher.

Finally, to perform an interaction analysis to determine which residues are most likely at the interface, the -is or --interaction-selection options can be used with an additional file giving the residue numbers of the receptor and ligand for which the interaction analysis will be performed. The input file consists of two lines, where the two lines are a space separated sequence of residue numbers for the receptor and ligand, respectively. As the interaction analysis is particularly expensive, only complexes consistent with all, or all - 1 are considered. To lower the barrier, the -ic option can again be used. Also limit the selected residues to only the solvent accessible ones. A simple example for the selection-file

1 2 3 4
101 302 888

This will select residue numbers 1, 2, 3, and 4 for the receptor, and 101, 302, and 888 for the ligand.

Output

Standard disvis outputs 5 files:

• accessible_complexes.out: a text file containing the number of complexes consistent with a number of restraints.
• violations.out: a text file showing how often a specific restraint is violated for each number of consistent restraints. This helps in identifying which restraint is most likely a false-positive if any.
• accessible_interaction_space.mrc: a density file in MRC format. The density represents the center of mass of the scanning chain conforming to the maximum found consistent restraints at every position in space. The density can be inspected by opening it together with the fixed chain in a molecular viewer (UCSF Chimera is recommended for its easier manipulation of density data, but also PyMol works).
• z-score.out: a text file giving the Z-score for each restraint. The higher the score, the more likely the restraint is a false-positive. Z-scores above 1.0 are explicitly mentioned in the output.
• disvis.log: a log file showing all the parameters used, together with date and time indications.

If an occupancy and/or interaction analysis was requested, disvis also outputs the following files:

• occupancy_N.mrc: a volume file giving a normalized indication of how frequent a grid point is occupied by the ligand, where N indicates the minimal required number of consistent restraints that resulted in the occupancy grid.
• [receptor|ligand]_interactions.txt: a text file containing the average number of interactions per complex each selected residue made, for both the receptor and ligand molecule. Each column denotes the minimal number of consistent restraints of each complex for which interactions were counted.

To visualize the most contacted residues, write the average number of interactions into the B-factor column of the receptor or ligand PDB file

python scripts/interactions_to_bfactor.py receptor.pdb receptor_interactions.txt N

where N is one of the column numbers of the interaction file. This writes receptor_interactions.pdb, which can be colored on B-factor, e.g. with spectrum b, blue_red in PyMOL or color bfactor in ChimeraX.

Licensing

If this software was useful to your research please cite us

Van Zundert, G.C.P. and Bonvin, A.M.J.J. (2015) DisVis: Visualizing and quantifying accessible interaction space of distance restrained biomolecular complexes. Bioinformatics 31, 3222-3224.

The software itself can be cited with its Zenodo DOI, 10.5281/zenodo.7775093, and CITATION.cff gives both references.

Apache License Version 2.0

The elements.py module is licenced under the MIT License, Copyright Christoph Gohlke.

Tested platforms

The GPU version has been tested on: NVIDIA GeForce GTX 1080 Ti and RTX 6000 Ada for Linux, and Apple M2 Pro for macOS (and, with the earlier Python 2 version, NVIDIA GeForce GTX 680 and AMD Radeon HD 7730M for Linux)

Reference results and timings for the test cases, on CPUs and GPUs, are in test-cases/README.md.

Contributing

See CONTRIBUTING.md for the development setup, code style, tests and pull requests.

AI Declaration

Large language models (LLM) were used in this repository since v3.0.0, for the port to Python 3 and the later changes. The changes were always reviewed by a human before being merged. See aidecl.yaml for machine-readable details of the LLM usage.