Official Resources
- Homepage: https://software.rcc.uchicago.edu/raptor/home.php
- Source Repository: https://github.com/uchicago-voth/raptor
- Documentation: https://github.com/uchicago-voth/raptor
- License: Open-source for non-profit use (Non-Profit Licensing Agreement)
Overview
RAPTOR (Rapid Approach for Proton Transport and Other Reactions) is an open-source software package for molecular dynamics simulations of chemical reactions in condensed phase. Developed by the research group of Prof. Gregory A. Voth at the University of Chicago, RAPTOR implements the Multiscale Reactive Molecular Dynamics (MS-RMD) methodology, which faithfully emulates reactive electronic structure through dynamic bonding at a fraction of the computational cost of ab initio methods.
RAPTOR is implemented as a plug-in package to LAMMPS, the popular molecular dynamics code from Sandia National Laboratory. This integration makes RAPTOR compatible with most LAMMPS features and functions, enabling large-scale simulations over hundreds of CPU cores via MPI. The package can also be interfaced with enhanced sampling methods to drive simulations toward the analysis of reactive rare events, with specialized collective variables (CVs) developed for this purpose. Recent advances include GPU acceleration and novel CVs for water wire formation modeling.
Scientific domain: Reactive molecular dynamics, proton transport, condensed phase chemistry
Target user community: Computational chemists studying reactive processes in solution and biomolecules
Theoretical Methods
- Multiscale Reactive Molecular Dynamics (MS-RMD)
- Multi-state empirical valence bond (MS-EVB) methodology
- Dynamic bonding for reactive topology changes
- Enhanced sampling with specialized collective variables
- Proton transport simulation
- Water wire formation modeling
- GPU-accelerated force evaluation
Capabilities (CRITICAL)
- Reactive MD simulations with dynamic bonding
- MS-RMD and MS-EVB methodology implementation
- Proton transport and other chemical reactions
- LAMMPS plug-in (compatible with most LAMMPS features)
- MPI parallelization over hundreds of CPU cores
- GPU acceleration for force evaluation
- Enhanced sampling collective variables for reactive events
- Water wire formation CVs
- Multi-proton simulations (SCI-MS-EVB)
- Compatible with LAMMPS 3Nov2022 release and newer
Inputs & Outputs
Input formats:
- LAMMPS input scripts with RAPTOR package commands
- MS-RMD parameter files
- Initial molecular configurations (LAMMPS data files)
Output data types:
- Molecular dynamics trajectories
- Reactive event statistics
- Proton hopping trajectories
- Energy and force outputs
- Collective variable time series
Interfaces & Ecosystem
- Programming language: C++ (LAMMPS plug-in)
- Base MD code: LAMMPS (required)
- Parallel computing: MPI for CPU, GPU acceleration
- Enhanced sampling: Interface with LAMMPS enhanced sampling methods
- Platform: Linux (HPC clusters)
Limitations & Known Constraints
- Requires LAMMPS installation and compilation
- Requires Non-Profit Licensing Agreement with developers
- MS-RMD parameterization needed for each new reactive system
- Source code only distribution (must compile within LAMMPS)
- Focus primarily on proton transport and similar reactions
Performance Characteristics
- MPI parallel scaling over hundreds of CPU cores
- GPU acceleration provides significant speedup (measured speedup vs MPI-only)
- MS-RMD is orders of magnitude faster than ab initio MD
- Optimal GPU settings depend on atoms per processor ratio
- CPU scaling tested on Cascade Lake Intel CPUs
Comparison with Other Codes
- vs ReaxFF: ReaxFF uses bond-order potentials; RAPTOR uses MS-RMD with explicit electronic states
- vs CP2K (QM/MM): CP2K uses ab initio QM region; RAPTOR uses reactive MD at classical cost
- vs LAMMPS (standard): Standard LAMMPS has fixed bonding; RAPTOR enables dynamic/reactive bonding
- vs EVB codes: RAPTOR implements the most advanced MS-EVB methodology
Best Practices
- Compile within LAMMPS following installation instructions carefully
- Use GPU acceleration for production simulations
- Interface with enhanced sampling for rare event analysis
- Use appropriate CVs for proton transport (water wire formation)
- Validate MS-RMD parameters against ab initio reference data
- Use SCI-MS-EVB for multi-proton systems
Verification & Sources
Primary sources:
- RAPTOR website: https://software.rcc.uchicago.edu/raptor/home.php
- GitHub: https://github.com/uchicago-voth/raptor
- S. Kaiser et al., J. Phys. Chem. B 128, 4959-4974 (2024)
Confidence: VERIFIED - Official website, GitHub repository, and peer-reviewed publication confirmed