PyRosetta Notebook Experiments

Exploratory notebooks for teaching and prototyping PyRosetta workflows: membrane protein modeling, enzyme design, symmetry setup, active-site analysis, structural alignment, and early ideas for a full online book chapter.

This repository is best read as a curated lab notebook. Some examples are polished chapter prototypes; others preserve useful experiments from older PyRosetta bindings so the reasoning and implementation details remain available.

Featured Work

Membrane Proteins: Moving Metals

The membrane-protein notebook demonstrates a visual workflow for opening a membrane protein in PyRosetta, introducing a metal, and sampling nearby residues with repacking over a sequence of steps.

• Notebook: Membrane 1 - Moving Metals.ipynb
• Supporting files: 1bl8.span, 1bl8.nometal.pdb

Prototype Book Chapter: Writing Protein Design Algorithms

Histidine tautomer example	Alternate tautomer example

The prototype chapter explores how to break down Rosetta packing behavior into readable data tables, then uses that visibility to build and debug a simple protein-design algorithm. It is the clearest example of the larger book format this repository was originally exploring.

• Notebook: Prototype of PyRosetta Jupyter Notebook.ipynb
• Section notes: prototype/README.md

Notebook Catalog

Section	Status	What it covers
membrane_proteins_moving_metals	Featured, PyRosetta 4	Membrane setup, metal placement, residue repacking, and animated sampling output.
prototype	Featured prototype	A book-chapter style walkthrough for extracting energies into dataframes and writing a simple design algorithm.
enzymedesign_in_pyrosetta	Working example	Enzyme-design setup in PyRosetta using Bagel/Foldit supporting files.
activesiteenergycalc	Archival example	Active-site RMSD and energy calculations against a reference structure.
poses_and_dataframes	Archival example, older bindings	Accessing pose energies with pandas and comparing Rosetta/PyRosetta scoring.
symmetry	Archival example, older bindings	Building symmetry setup logic directly in PyRosetta.
getting_tmalign_to_work_w_ligands	Archival example, older bindings	Using TMalign-style superposition for proteins with different sequences.

Repository Layout

.
|-- README.md
|-- resources/
|   |-- images/                 # Shared README and notebook visuals
|   `-- protein_structures/      # Shared structure inputs
|-- membrane_proteins_moving_metals/
|-- prototype/
|-- enzymedesign_in_pyrosetta/
|-- activesiteenergycalc/
|-- poses_and_dataframes/
|-- symmetry/
|-- getting_tmalign_to_work_w_ligands/
`-- sjb_util.py                  # Helper utilities used by some legacy demos

Running The Notebooks

These notebooks were written across multiple PyRosetta eras. The featured membrane and prototype notebooks expect PyRosetta 4-era APIs, while several archival examples use older bindings and may need small updates before running on a modern installation.

General setup:

1. Install PyRosetta using the license and platform-specific instructions from RosettaCommons.
2. Create a Python environment with Jupyter, pandas, matplotlib, and seaborn.
3. Launch Jupyter from the repository root so relative paths to PDB, params, span, image, and XML files resolve correctly.

Relationship To The Official PyRosetta Notebooks

A later RosettaCommons effort produced the broad PyRosetta notebook collection this project originally hoped would exist. This repository now serves as a smaller companion archive of experiments, chapter prototypes, and teaching ideas.

• Official notebooks: https://github.com/RosettaCommons/PyRosetta.notebooks
• Associated preprint: https://www.preprints.org/manuscript/202002.0097/v1

Notes

• PyRosetta and Rosetta are distributed under RosettaCommons licensing terms; this repository does not include PyRosetta itself.
• sjb_util.py is retained for examples that depend on it, but it has not been cleaned up as a public API.
• Data files are included only where needed to make individual notebooks easier to inspect and reproduce.