PyL3dMD

PyL3dMD calculates three-dimensional (3D) molecular descriptors from molecular dynamics (MD) trajectories, including LAMMPS output, to characterize dynamic molecular properties for cheminformatics, QSPR, and materials modeling.

Key Features:

• Extensive Descriptor Calculation: Computes over 2000 distinct 3D molecular descriptors that quantify biological, physical, and chemical properties.
• LAMMPS Compatibility: Processes atomic trajectories generated by LAMMPS MD simulations.
• Performance Efficiency: Demonstrated efficient computation for large and complex molecular systems over extended simulation durations in benchmark studies.
• Operating-Condition Sensitivity: Derives descriptors that capture static properties and effects of operating conditions such as temperature and pressure.
• Data for QSPR and Machine Learning: Produces descriptor sets suitable for quantitative-structure-property-relationship (QSPR) models and machine learning workflows.

Scientific Applications:

• Cheminformatics: Extraction of 3D descriptors from MD trajectories to analyze molecular interactions and dynamic behavior.
• Materials Design: Characterization of molecular systems under varying conditions to inform materials development.
• Predictive Modeling: Generation of descriptor inputs for QSPR and machine learning-based predictive models.

Methodology:

PyL3dMD applies post-processing routines to MD simulation trajectories (including LAMMPS output) to derive 3D molecular descriptors that capture static properties and effects of operating conditions such as temperature and pressure.