ULaMDyn

ULaMDyn automates the unsupervised analysis of large datasets generated by nonadiabatic molecular dynamics (NAMD) simulations to uncover hidden patterns in molecular trajectories and enable intuitive understanding of excited-state processes.

Key Features:

• Dimensionality reduction: Employs advanced dimensionality reduction techniques to address the high dimensionality and complexity of NAMD data.
• Clustering: Uses clustering techniques to identify hidden patterns and group similar molecular geometries in NAMD trajectories.
• Newton-X integration: Integrates seamlessly with the Newton-X platform for NAMD simulation analysis.
• Scalable analysis: Provides a streamlined, scalable solution for interpreting large NAMD datasets.

Scientific Applications:

• Excited-state dynamics analysis: Enables interpretation of excited-state processes in molecular systems through automated pattern recognition in NAMD trajectories.
• Critical geometry identification: Efficiently identifies critical molecular geometries and critical nonadiabatic transitions in photochemical dynamics, as demonstrated for fulvene.

Methodology:

ULaMDyn employs dimensionality reduction and clustering techniques for unsupervised analysis of NAMD trajectory data.