ConfoState documentation

ConfoState classifies membrane protein structures into conformational states described in the literature.

Contents

• ConfoState — Usage Guide
  • 1. Set up
  • 2. Download structures
  • 3. Annotate structures
  • 4. Extract features
  • 5. Train and evaluate a classifier
  • Notes
• Development guide
  • Style
  • Pre-commit
• download_structures
  • Synopsis
  • Description
  • Options
  • Examples
  • Output
  • Notes
• extract_feature_vectors
  • Synopsis
  • Description
  • Options
  • Examples
  • Output
• LeuT descriptors
  • Macromolecular Conformations in the LeuT Family
  • Curated List of Descriptors
  • References
• Feature Extraction
  • Package layout
  • Quick start
  • Data provenance & current status (Person 3)
  • API
  • Feature modules
  • Feature column manifest
  • Dependencies on Person 1 (data)
  • Testing
  • Future work
• Explainability
  • Package layout
  • Quick start
  • Importance methods
  • Citations
  • Integration with Person 5 (CLI)
  • Testing
  • Future work
• API reference
  • extract_features()
  • evaluate_model()
  • get_baseline_models()
  • get_registered_model()
  • load_annotations()
  • register_model()
  • run_training()
  • write_evaluation_report()
  • load_annotations()
  • load_from_input_dir()

Quick start

pip install -e .

See ConfoState — Usage Guide for workflows, API reference for the Python API, and Development guide for coding style and pre-commit.