Pipeline Overview

The PyGVAMP analysis pipeline processes molecular dynamics trajectory data through several stages:

1. Data Loading — Read trajectory files and topology information
2. Graph Construction — Build molecular graphs from atomic coordinates
3. Training — Train the GraphVAMPNet model to learn state representations
4. Analysis — Extract conformational states, transition kinetics, and attention patterns
5. Visualization — Generate interactive HTML reports

For execution details, see Pipeline Execution.

For planned improvements, see the Pipeline Roadmap.