ATOMDANCE 2.0 – software/example data/example results – statistical machine learning for comparative protein dynamics

This is a data/results repository (atomdance_results_repo.zip) for the examples offered with software ATOMDANCE 2.0, a python GUI pipeline for the statistical machine learning post-processing of comparative protein dynamics. The software is available at

GitHub (https://github.com/gbabbitt/ATOMDANCE-comparative-protein-dynamics)

GitHub pages (https://gbabbitt.github.io/ATOMDANCE-comparative-protein-dynamics/)

The software is also copied into the .zip file here.

ATOMDANCE 2.0 offers divergence metrics and machine learning methods for amino acid site-wise comparison of molecular dynamics simulations conducted in on a protein in two functional states (e.g. bound vs. unbound, wildtype vs. mutant, hot vs cold temperature, etc. )  ATOMDANCE also offers methods for identifying dynamics that are coordinated across amino acid sites, and for identifying non-neutral evolutionary changes in dynamics over deep time. For more information see the documentation in the .zip or at the websites above. ATOMDANCE 2.0 was developed in 2023 by Dr. Gregory A. Babbitt, Dr. Ernest P. Fokoue and students at the Rochester Institute of Technology and is offered freely without guarantee under GPL 3.0 license.  

In this .zip file, we offer .pdb, ,prmtop, and .nc files (i.e. structure, topology, and 1ns trajectory) files in the 1ns examples folder.  These correspond to all the examples shown in the software documentation on GitHub.  Folders for output results include one negative control where ubiquitin(PDB 1ubq) is compared using two simulations conducted in an identical state, and 3 positive controls including (A) TATA binding protein compared in its DNA-bound and unbound state, (B) BRAF kinase compared in a drug-bound and unbound state (drug-sorafenib). and (C) the viral SARS-CoV-2 receptor binding domain and its human target protein angiotensin converting enzyme 2 (ACE2) compared in bound vs unbound state.  The negative and positive control 1 folders also contain results from analyses to identify amino acid sites with coordinated dynamics and results to identify non-neutral evolution of dynamics across distant evolutionary orthologs. See documentation PDF for more details.

A GUI for generating molecular dynamics simulations via opensource AmberTools and openMM is also available here (MDgui.py)

A movie maker feature (makeMovie.py) also allows for movies of functional protein dynamics  Movies for the positive controls are in the folder ‘movies’ also contained in the .zip

please cite us (as well as ChimeraX and cpptraj)

preprint at https://www.biorxiv.org/content/10.1101/2023.04.20.537698v2

Babbitt G.A. Coppola E.E. Mortensen J.S. Adams L.E. Liao J. K. 2018. DROIDS 1.2 – a GUI-based pipeline for GPU-accelerated comparative protein dynamics. BIOPHYSICAL JOURNAL 114: 1009-1017. CELL Press.

Babbitt G.A. Fokoue E. Evans J.R. Diller K.I. Adams L.E. 2020. DROIDS 3.0 - Detection of genetic and drug class variant impact on conserved protein binding dynamics. BIOPHYSICAL JOURNAL 118: 541-551 CELL Press.