Structure Visualization Manager

GitHub Link to Code.

Structure visualization manager for molecular structure analysis.

This manager coordinates structure visualization workflows through specialized services for feature importance and feature-based approaches.

class mdxplain.structure_visualization.manager.structure_visualization_manager.StructureVisualizationManager(use_memmap: bool = False, chunk_size: int = 2000, cache_dir: str = './cache')

Manager for structure visualization workflows.

Provides two specialized services:

feature_importance: Visualizations based on feature importance analysis
feature: Visualizations based on feature and data selectors

Warning

When using PipelineManager, the pipeline_data parameter is automatically injected. Do NOT provide it manually.

Examples

>>> # Feature importance approach
>>> pipeline.structure_visualization.feature_importance.create_pdb_with_beta_factors(
...     "my_viz", "dt_analysis", n_top=10
... )
>>> pipeline.structure_visualization.feature_importance.visualize_nglview_jupyter(
...     "my_viz", n_top_global=3
... )

>>> # Feature selector approach
>>> pipeline.structure_visualization.feature.create_representative_pdbs(
...     "my_viz",
...     data_selectors=["cluster_0", "cluster_1"],
...     selector_centroid="coords_all",
...     selector_features="distances"
... )
>>> pipeline.structure_visualization.feature.visualize_nglview_jupyter("my_viz")

__init__(use_memmap: bool = False, chunk_size: int = 2000, cache_dir: str = './cache')

Initialize structure visualization manager.

Creates manager instance and sets up dedicated output directory for structure visualization files (PDBs, scripts, etc.).

Parameters

use_memmapbool, default=False: Whether to use memory mapping for large datasets
chunk_sizeint, default=2000: Chunk size for processing large arrays
cache_dirstr, default=”./cache”: Base cache directory. Manager creates subdirectory “structure_viz” as dedicated output directory.

Returns

None: Initializes manager instance with output directory

Notes

Output directory is set to cache_dir/structure_viz and created automatically during initialization.

property feature_importance

Service for feature-importance-based visualization.

Provides methods for creating and visualizing molecular structures with beta-factors derived from feature importance analysis.

Returns

StructureVizFeatureImportanceService: Service instance for FI-based visualization

Examples

>>> pipeline.structure_visualization.feature_importance.create_pdb_with_beta_factors(
...     "my_viz", "dt_analysis", n_top=10
... )

property feature

Service for feature-selector-based visualization.

Provides methods for creating and visualizing molecular structures based on feature selectors and data selectors without requiring feature importance analysis.

Returns

StructureVizFeatureService: Service instance for feature-based visualization

Examples

>>> pipeline.structure_visualization.feature.create_representative_pdbs(
...     "my_viz",
...     data_selectors=["cluster_0", "cluster_1"],
...     selector_centroid="coords_all",
...     selector_features="distances"
... )