molecular-simulations¶
A comprehensive Python toolkit for building, running, and analyzing molecular dynamics simulations using the AMBER force field ecosystem and OpenMM.
Features¶
- System Building
Build explicitly solvated systems with OPC water, implicit solvent systems, and parameterize small molecules via GAFF2.
- Simulation Engine
Run simulations with OpenMM (v8.0+) with GPU acceleration. Supports advanced methods like constant-pH MD and Empirical Valence Bond simulations. Deploy to HPC clusters via Parsl.
- Analysis Tools
Automatic clustering with KMeans++, protein-protein interaction analysis, interaction energy fingerprinting, MM-PBSA binding free energy calculations, SASA calculations, and interface scoring (ipTM, ipSAE, pDockQ).
Quick Example¶
from molecular_simulations.build import ExplicitSolvent
from molecular_simulations.simulate import Simulator
from molecular_simulations.analysis import Fingerprinter
from pathlib import Path
# Build a solvated system
builder = ExplicitSolvent(Path("./outputs"), Path("protein.pdb"))
builder.build()
# Run simulation
sim = Simulator(builder.out.parent)
sim.run()
# Analyze interactions
fp = Fingerprinter("system.prmtop", trajectory="prod.dcd", target_selection="segid A")
fp.run()
fp.save()
User Guide
API Reference
Development