atomworks documentation

Welcome to atomworks — a toolkit for converting, parsing, and manipulating biological structure and sequence data, inspired by the Biotite library. Quickly convert between formats, extract features, and prepare data for machine learning or structural analysis.

Navigation

• Tutorial
  • Getting Started
• How to Build a Model with AtomWorks
  • Part 1: Prepare the Data
  • Part 2: Creating a Transform Pipeline
  • Part 3: Write a Model
  • Part 4: Train a Model
  • Scripts
• Cookbook
  • Input
  • Output
  • Structure Manipulation
  • Coordinate Manipulation
  • Calculations
  • Visualization
  • Other
• Installation
  • 0. Prerequisites
  • 1. Installing via pip (recommended)
  • 2. Development Installation
  • 3. Running the Test Suite
  • 4. Setting Up Full PDB/CCD Mirrors
• Glossary
  • Entities vs. Instances
  • Suffixes
  • Chains, PN_Units, and Molecules
  • Combinatorial Nomenclature - Worked Example
• API
  • Core Modules
  • IO
  • ML
• Examples
• Contributing
  • As you code
  • As you commit
  • As you finalize a PR
  • As you review
  • Contributing to the documentation
  • Other Resources
  • Publishing a release
• Data Mirrors
  • Setting Up Mirrors
• Training on the PDB
  • Step 1 — Mirror the PDB (mmCIFs)
  • Step 2 — Get PDB metadata (PN units and interfaces)
  • Step 3 — Configure an AF3-style dataset (example: train only on D-polypeptides)
  • Step 4 — Train a model
• Hugging Face dataset repositories
• Multiple Sequence Alignment in AtomWorks
  • Find
  • Filter
  • Generate
  • Organize