Protviz: Protein Annotation Visualiser
Protviz is a Python package designed to retrieve and visualize various protein annotations and structural information. It allows users to fetch data from multiple bioinformatics databases and plot this information along a protein sequence using a flexible track-based system.
Features
- Data Retrieval:
- Fetch protein sequence length from UniProt.
- Retrieve PDB coverage and ligand interaction data from PDBe.
- Get TED domain annotations from the TED database.
- Fetch pLDDT scores and AlphaMissense data from the AlphaFold Database (AFDB).
- Track-Based Visualization:
- AxisTrack: Displays the sequence axis with tick marks.
- PDBTrack: Shows PDB structure coverage, with options to display as individual entries or a collapsed overview.
- LigandInteractionTrack: Visualizes ligand binding sites on the protein.
- TEDDomainsTrack: Displays TED domain annotations.
- AlphaFoldTrack: Shows AlphaFold prediction metrics like pLDDT and average AlphaMissense pathogenicity scores.
- CustomTrack: Allows plotting of arbitrary user-defined annotations (ranges or points) with customizable labels and colors.
- Core Plotting Functionality:
- Combines multiple tracks into a single, coherent plot.
- Supports zooming into specific regions of the protein sequence.
- Option to save plots to a file.