---

Highlights

• Perform complex molecular dynamics analyses with intuitive, single-line commands
• Automatically generate slide-ready, publication-quality figures with consistent axes, fonts, and colorbars
• Seamlessly switch between Python API for advanced workflows and CLI for rapid batch processing
• Scalable workflows that handle everything from quick exploratory analysis to large-scale production runs

---

Analysis Modules

Analysis	Description
rmsd	Root-Mean-Square Deviation relative to a reference frame
rmsf	Per-atom Root-Mean-Square Fluctuation
rg	Radius of Gyration for molecular compactness
hbonds	Hydrogen bond detection and count using Baker-Hubbard algorithm
ss	Secondary Structure assignments using DSSP
cluster	Trajectory clustering using KMeans, DBSCAN, and Hierarchical methods
sasa	Solvent Accessible Surface Area with total, per-residue, and average per-residue
dimred	Dimensionality reduction using PCA, MDS, and t-SNE methods

---

Installation

We strongly recommend installing FastMDAnalysis in a virtual environment to avoid conflicts with system packages and ensure the fastmda command is available in your PATH.

Install from conda-forge (Recommended)

conda create -n fastmda_env python=3.11

conda activate fastmda_env

conda install -c conda-forge fastmdanalysis

Verify installation

fastmda analyze -h

[Alternatively] Install from PyPI

Make sure you have Python 3.9+ installed and available as 'python'.

python -m venv fastmda_env

# On Linux/macOS:
source fastmda_env/bin/activate

# Or on Windows:
# fastmda_env\Scripts\activate

pip install fastmdanalysis

Verify installation

fastmda analyze -h

---

Usage

Command-Line Interface (CLI)

After installation, you can run FastMDAnalysis from the command line using the fastmda command. Global options allow you to specify the trajectory and topology file paths. Optionally, specify frame selection and atom selection. Frame selection is provided as a tuple (start, stop, stride). Negative indices (e.g., -1 for the last frame) are supported. If no options are provided, the entire trajectory and all atoms are used by default.

Run the analyze orchestrator to execute multiple analyses in one go.

Run all available analyses

fastmda analyze -traj path/to/trajectory -top path/to/topology

Include specific analyses

fastmda analyze -traj traj.dcd -top top.pdb --include rmsd rg

Exclude specific analyses

fastmda analyze -traj traj.dcd -top top.pdb --exclude sasa dimred cluster

Supply options via file (YAML or JSON)

fastmda analyze -traj traj.dcd -top top.pdb --options options.yaml

Create a slide deck from generated figures

fastmda analyze -traj traj.dcd -top top.pdb  --slides

Global flags:

• --frames start,stop,stride (e.g., 0,-1,10)
• --atoms "MDTraj selection" (e.g., "protein and name CA")
• --output DIR (output directory name)
• --verbose (prints progress and writes logs under <command>_output/ unless --output is set)

Show help:

• fastmda -h
• fastmda analyze -h

Options file (schema)

Provide per-analysis keyword arguments in a single file. CLI and Python API share the same schema:

# options.yaml
rmsd:
  ref: 0
cluster:
  methods: [kmeans, hierarchical]
  n_clusters: 5

Slides:

• --slides creates fastmda_slides_<ddmmyy.HHMM>.pptx in the current working directory.
• --slides path/to/deck.pptx writes to an explicit filename.

Python API

Instantiate a FastMDAnalysis object with your trajectory and topology file paths.

Run the analyze orchestrator to execute all available analyses.

from fastmdanalysis import FastMDAnalysis
from fastmdanalysis.datasets import TrpCage  # optional helper

fastmda = FastMDAnalysis(TrpCage.traj, TrpCage.top)
fastmda.analyze()

Include or Exclude specific analyses; specify options, generate slides

fastmda = FastMDAnalysis(TrpCage.traj, TrpCage.top)
result = fastmda.analyze(
    include=["rmsd", "rg"],                 # or exclude=[...]; omit to run all
    options={"rmsd": {"ref": 0, "align": True}},
    slides=True                             # or slides="results.pptx"
)

Notes

• Figures are saved during each analysis; slide decks include all figures produced in the run.
• MDTraj may emit benign warnings (e.g., dummy CRYST1 records); they do not affect results.

---

Output

Output includes data tables, figures (already stylized for slides), slide deck, log file ...

---

Validation and Benchmarks

• To validate FastMDAnalysis go to https://github.com/aai-research-lab/FastMDAnalysis/blob/validation/README.md
• For benchmarks go to https://github.com/aai-research-lab/FastMDAnalysis/blob/benchmark/README.md

---

Documentation

The documentation [under development] (with an extensive User Guide) is available here.

---

Contributing

Contributions are welcome. Please submit a Pull Request.

Development Installation

If you want to contribute or modify the code:

# Clone the repository
git clone https://github.com/aai-research-lab/FastMDAnalysis.git
cd FastMDAnalysis

# Create and activate virtual environment
python -m venv fastmda_env
source fastmda_env/bin/activate  # On Windows: fastmda_env\Scripts\activate

# Install in development mode with test dependencies
pip install -e ".[test]"

# Verify installation
fastmda -h
fastmda analyze -h

---

Citation

If you use FastMDAnalysis in your work, please cite:

Aina, A. and Kwan, D. (2025) “FastMDAnalysis: Software for Automated Analysis of Molecular Dynamics Trajectories”. ChemRxiv. https://doi.org/10.26434/chemrxiv-2025-x8xnq

@article{fastmdanalysis,
  author       = {Adekunle Aina and Derrick Kwan},
  title        = {FastMDAnalysis: Software for Automated Analysis of Molecular Dynamics Trajectories},
  year         = {2025},
  doi          = {10.26434/chemrxiv-2025-x8xnq},
  publisher    = {ChemRxiv},
  url          = {https://doi.org/10.26434/chemrxiv-2025-x8xnq},
  type         = {preprint}
}

---

License

FastMDAnalysis is licensed under the MIT license.

---

Acknowledgements

FastMDAnalysis builds upon excellent open-source libraries to provide its high-performance analysis capabilities and to improve workflow efficiency, usability, and reproducibility in molecular dynamics trajectory analysis. We gratefully acknowledge:

• MDTraj for foundational trajectory I/O and analysis modules
• NumPy/SciPy for efficient numerical computations
• scikit-learn for advanced machine learning algorithms
• Matplotlib for publication-quality visualization

While leveraging these robust tools, FastMDAnalysis streamlines analysis for students, professionals, and researchers, especially those new to molecular dynamics. We thank the scientific Python community for their contributions to the ecosystem.