Repository accompanying the paper:

'Characterizing the kinetic and structural ensembles of intrinsically disordered proteins using writhe' By Thomas Sisk, Simon Olsson and Paul Robustelli.

preprint: https://www.biorxiv.org/content/10.1101/2025.04.26.650781v1

---

Results

• Trajectory Analysis and Dynamic Modeling

  • Demonstrations of the code to reproduce (or load and visualive) trajectory analysis and MSM results from the manscript are presented via jupyter notebooks 👉 traj_analysis_results.

• Generative Modeling

  • Analysis of the samples generated from score-based DDPMs trained using the PaiNN and 'Writhe-PaiNN' architectures and scripts to train and sample either model 👉 deep_generative_models.

---

Data

The raw data pertaining to our results is available at: LINK TO DATA REPOSITORY

• The data archive includes:
  • precomputed writhe features for segment lengths 1 to 5.
  • tCCA decompositions and VAMP2 scores for all writhe datasets considered in the paper.
  • Kmeans clusters estimated over a range of parameters.

Software - writhe_tools

• Accompanying this project is an open-source python software package hosted and archived on the Python Package Index, PyPI : https://pypi.org/project/writhe-tools/

• The package can be installed via:

pip install writhe-tools

• Tutorial on how to use writhe_tools and the algorithm used to compute the writhe demonstrated through example analysis of ACTR 👉 traj_analysis_results

• The writhe_tools package will be actively maintained at https://github.com/tommysisk/writhe_tools and new releases will be distributed on PyPI until further notice. Please direct all purely package related inquires accordingly. writhe_tools version 0.0.18 should be utilized when reproducing demonstrations of the code in this repository and is permenantly available via archive at https://pypi.org/project/writhe-tools . Future releases are not guarunteed to work with the sample code provided in this repository.