StACKER

Stacking Analysis and Conformational Kinetics for Examining Residues

Analyzes pi-stacking interactions between residues in Molecular Dynamics (MD) trajectories.

Developed by Eric Sakkas (esakkas@wesleyan.edu) in the Weir Lab at Wesleyan University.

Runs on Python 3.10.x+ and has the following dependencies: mdtraj, pandas, NumPy, matplotlib, seaborn

StACKER Documentation is available here: https://esakkas24.github.io/

Overview

Manipulates the outputs of an MD simulation and analyzes the pi-stacking interactions. Creates a "Pi-Stacking Fingerprint" for a structure at each frame. Presents Pi-stacking interactions between two residues through analysis of their relevant movement.

Installation Instructions

StACKER can be installed through GitHub or through PyPi:

Install StACKER with pip

In the command line, run:

pip install pistacker

This will install StACKER, activate the command line option stacker, and install all the necessary dependencies.

Clone StACKER repository to local computer

In the command line, run:

git clone https://github.com/esakkas24/stacker.git

Descend into the directory and download the neccessary dependencies:

cd stacker
pip install -r requirements.txt
pip install setuptools
python setup.py install

This will install StACKER, activate the command line option stacker, and install all the necessary dependencies.

If you need to download any of the dependencies individually

Download mdtraj

pip3 install mdtraj

If installing mdtraj presents issues on the newest version of pip, run the script get-pip.py to download an older version of pip:

python3 installation/get-pip.py

The output may come with a warning showing the location of the new pip version:

WARNING: The scripts pip, pip3, and pip3.8 are installed in '/Users/ericsakkas/Library/Python/3.8/bin' which is not on PATH

If it does, use this new path to install mdtraj:

/Users/ericsakkas/Library/Python/3.8/bin/pip3 install mdtraj

Else, use the usual pip3 install (as shown in installation.mp4)

If successful, this will also install the NumPy dependency:

Successfully installed astunparse-1.6.3 mdtraj-1.9.9 numpy-1.24.4 ...

Install Pandas

pip3 install pandas

Install matplotlib

pip3 install matplotlib

Install seaborn

pip3 install seaborn

Install sklearn

pip3 install scikit-learn

Testing Features

All features can be tested by running the unit tests at the end of each Python script, or by running stacker.py in the command line. All tests are explained in the testing/testing.md file.

MD Files are provided for testing convenience in the testing folder:

• first10_5JUP_N2_tUAG_aCUA_+1GCU_nowat.mdcrd : A 10-frame trajectory file
• 5JUP_N2_tUAG_aCUA_+1GCU_nowat.prmtop : The associated Topology File with the above trajectory.
• 5JUP_N2_tUAG_aCUA_+1GCU_nowat_mdcrd_3200frames.pdb : A larger Trajectory combined with a Topology file with 3200 frames.

Future Features

• Usage for more trajectory types beyond mdcrd prmtop and pdbs

Features

1. Command Line Interface to run stacker commands
2. A Vector Class to compute distances within the 3D space of the MD simulation.
3. Users can convert the .trj output of an MD simulation (which contains atom position, velocities, and forces per frame with no info on atom identity) to a .pdb file (which has atom idenity, position, the residue they make up, and more).
4. Users can input an MD simulation and two residues and get a map of how those two residues move relative to each other (Figure D shows a heatmap of how one residue moves in the perspective of another).
5. Users can get a "Stacking Fingerprint" for a given frame of a structure. A Stacking Fingerprint involves checking every pair of residues for stacking interactions between the two residues, and returning a pairwise comparison as a Matrix.
6. A visualization interface that allows for the display of residue-residue movement from Feature 4 (as in Figure D) and the display of a heatmap for stacking interactions as in Feature 5
   • Heatmap: matrix where x-axis and y-axis are residue idenities within the strucutre (eg. residue 1, residue 48, etc.) and matrix(i,j) is colored by distance between the residues, where stacking interactions occur at around 3-4 Angstroms apart.

Stakeholders and Intended Users

The package is intended to be used by anyone in academic or computational biology centers running molecular dynamics. No explicit prerequisites, but a conceptual understanding of MD output files is beneficial. The stakeholders include the users, other researchers reliant on the computational data, and any beneficiaries of the overall research conducted using StACKER.