Pre/Post Processing Tools

Materials Design, Inc. develops MedeA®, an atomistic simulation and modeling environment that provides productivity, model-building, and analysis tools for use with LAMMPS. MedeA® simplifies LAMMPS simulations with flowcharts that assemble complex protocols from discrete LAMMPS stages, which can be shared, edited, and reused. Atomistic models can be constructed with the collection of MedeA® builders, and validated methods predict mechanical properties, elastic constants, diffusivity, transport properties, and cohesive energies from LAMMPS simulations. MedeA® also manages force fields for organic, inorganic, and metallic systems.

OVITO is a visualization and analysis software for output data generated in molecular dynamics, atomistic Monte-Carlo, and other particle-based simulations. It is particularly well interfaced with LAMMPS, since some of the OVITO developers are frequent LAMMPS users. The OVITO Pro version can also load the LAMMPS shared library to create and run LAMMPS simulations directly from OVITO, loading the data without intermediate files.

Scienomics has developed an interface to LAMMPS as part of their Materials and Processes Simulations (MAPS) platform, which lets both novice and expert users quickly create LAMMPS input files for atomistic and DPD simulations, with full phone and email support. The LAMMPS plugin within MAPS also lets users visualize and analyze LAMMPS output, and MAPS offers a number of atomistic and coarse-grained force fields plus builder tools for creating input geometries. MAPS implements a complete Build–Simulate–Analyze workflow across an array of simulation engines including LAMMPS.

Atomify is a web-based application that runs LAMMPS simulations purely in the browser, with real-time visualization and plotting of thermodynamic quantities. It ships with several example simulations and lets you analyze them in a browser-based Jupyter notebook. Because it requires no installation, it is an excellent choice for new LAMMPS users and for teaching environments.

atomman is open-source software developed by Lucas Hale at NIST: a Python package for interacting with large-scale atomic systems that lets users prepare, run, and analyze MD simulations entirely from Python. It supports efficient calculations on millions of atoms with many per-atom properties, generation of defects (point defects, dislocation monopoles), built-in defect analysis (slip vector, Nye tensor), and direct calls to LAMMPS from Python. It converts to/from ASE and PyMatGen, loads CIF/POSCAR and LAMMPS atom/dump files, includes unit-conversion tools, and runs on Linux, Windows, and macOS.

LAMMPS-GUI is a cross-platform graphical text editor for LAMMPS input files with syntax highlighting, auto-completion, inline help, and indentation support. Written in C++ with the Qt framework, it calls LAMMPS directly through the library interface instead of launching an external executable, so it can display information while LAMMPS runs and show visualizations created by the dump image command. It is well suited to teaching beginners with consistent behavior across Linux, macOS, and Windows, and is integrated with a collection of LAMMPS tutorials. It is written and maintained by Axel Kohlmeyer.

LOOS (Lightweight Object-Oriented Structure analysis library) is a package-agnostic, open-source package for analyzing molecular dynamics simulations that runs on all major Linux distributions and macOS. It ships with roughly 150 pre-packaged analysis tools, from standard tasks (trajectory manipulation, principal component analysis) to novel ones (assessing convergence, measuring membrane properties), and is designed for rapid development of new analysis tools, particularly via its Python wrappers. It is available on GitHub.

LUNAR (LAMMPS Utility for Network Analysis and Reactivity) is a standalone Python (3.7+) toolkit that supplements LAMMPS, focused on pre- and post-processing of LAMMPS inputs and outputs with an emphasis on structure–property relationships for ICME process modeling of polymers — though many of its codes are useful outside process modeling. It was written by Josh Kemppainen (Michigan Technological University, advisor Dr. Gregory M. Odegard) and is maintained by Josh Kemppainen and Dr. Jacob R. Gissinger. All source is provided so users can modify it.

When using LUNAR, please cite “LUNAR: Automated Input Generation and Analysis for Reactive LAMMPS Simulations”, https://doi.org/10.1021/acs.jcim.4c00730.

freud is a Python library developed and supported by Sharon Glotzer’s group (University of Michigan) that provides a simple, flexible, powerful set of tools for analyzing trajectories from molecular dynamics or Monte Carlo simulations. High-performance, parallelized C++ computes standard tools such as radial distribution functions, correlation functions, order parameters, and clusters, as well as original methods including potentials of mean force and torque (PMFTs) and local-environment matching. freud supports many input formats and outputs NumPy arrays, integrating with the scientific Python ecosystem.

• Tutorial: g(r) analysis
• Documentation: https://freud.readthedocs.io

MDAnalysis is an object-oriented Python library to analyze trajectories from molecular dynamics simulations. It reads particle-based trajectories (including single coordinate frames such as biomolecules in PDB format) and exposes atomic coordinates as NumPy arrays, providing a flexible and relatively fast framework for complex analysis tasks. It implements powerful atom-selection commands, and trajectories can be manipulated (for example, fit to a reference structure) and written out.

MDANSE (Molecular Dynamics Analysis of Neutron Scattering Experiments) helps predict neutron observables from MD trajectories. Inelastic and quasi-elastic neutron scattering (INS and QENS) probe molecular dynamics in materials, and MDANSE implements the operations needed to compare simulations with experiment — velocity auto- and cross-correlation functions, Fourier transforms, and convolutions with instrument parameters. It interfaces with more than ten MD codes, including LAMMPS, CASTEP, VASP, Gromacs, CHARMM, and DFTB. The package is being modernized and is available on GitHub and PyPI.

mdapy (Molecular Dynamics Analysis with Python) offers a wide range of flexible, user-friendly tools for analyzing atomic trajectories from MD simulations across Windows, Linux, and macOS. It uses the TaiChi project to accelerate pure Python toward C++ performance and is optimized for parallel processing on multicore CPUs and GPUs. mdapy handles LAMMPS DUMP and DATA files, VASP POSCAR, universal XYZ, and CIF formats, and stores all data as NumPy ndarrays for seamless integration with the scientific Python ecosystem.

• Documentation: https://mdapy.readthedocs.io/
• Publication: https://doi.org/10.1016/j.cpc.2023.108764

MolTwister is an open-source package that aids in constructing molecular systems for MD simulations, but whose primary purpose is a collection of analysis tools for post-processing trajectories and data — density profiles, velocity autocorrelation functions, radial distribution functions, dihedral distributions, and more. It runs as an extension of the command line: commands it does not recognize are passed through to the regular shell, and a 3D view window displays the molecular system being edited or created. See the documentation.

PyLAT (Python LAMMPS Analysis Tools) is Python software developed in the group of Ed Maginn at Notre Dame for post-processing output from LAMMPS simulations. Details are in this publication.

TRAVIS (“TRajectory Analyzer and VISualizer”) is a free program package for analyzing and visualizing Monte Carlo and molecular dynamics trajectories. Its aim is to collect as many analyses as possible in one program, making it unnecessary to use many different tools to evaluate simulations. TRAVIS is written in C++, is open-source freeware under the GPLv3, is platform-independent with no external libraries, and is easy to install and use.

Developed and maintained by Pieter J. in ’t Veld (BASF). Enhanced Monte Carlo (EMC) provides an environment for creating and manipulating input structures for particle simulations using COMPASS, CHARMM, OPLS, Martini, DPD, or colloidal force fields. A scripting language manages access to its functionality: manipulation of molecular or coarse-grained structures through SMILES strings, typing those structures for selected force fields, and building conformations using Monte Carlo principles to unoverlap atoms. EMC provides output ports to LAMMPS, PDB, and XYZ formats; compiled versions for Linux, macOS, and Windows are available. EMC has a user forum on MatSci.org.

Examples of systems built with EMC (click for a larger image):

Moltemplate, developed and maintained by Andrew Jewett (UCSB), is distributed with LAMMPS in the tools/moltemplate directory. It was designed for building coarse-grained biomolecular models and can create both LAMMPS data files (geometry and topology) and LAMMPS input scripts (force fields, fixes, groups), giving users access to all of the force fields available in LAMMPS. Molecules are written in a compact, readable template format (.LT), can be shared, and can be copied, combined, and nested to build larger molecules.

Examples of systems built with Moltemplate (click for a larger image):

Developed and maintained by Axel Kohlmeyer (Temple U). TopoTools is a molecule builder that leverages VMD and Tcl to create LAMMPS data files and convert them to and from other formats. It has two components: a middleware script that extracts and manipulates topology information, and several high-level applications built on top of it (for example, to read/write data files and replicate and merge systems). Together with VMD, TopoTools can infer topology from PDB and PSF files and atom-pair distances, and solvate a protein.

Examples of systems built with TopoTools (click for a larger image):

InterMol is written in Python and performs Desmond ⇄ Gromacs ⇄ LAMMPS conversions natively. AMBER → X is carried out by converting AMBER to GROMACS and then to other programs using ParmEd; AMBER → CHARMM is done by ParmEd directly.

AMBER2LAMMPS is a Python utility to convert AMBER molecular dynamics files to the LAMMPS data format, with an enhanced command-line interface and error handling. It can also be loaded as a Python module to embed the conversion in more complex workflows, and it uses ParmEd.

Avogadro is an advanced molecule editor and visualizer designed for cross-platform use in computational chemistry, molecular modeling, bioinformatics, materials science, and related areas. It offers flexible, high-quality rendering and a powerful plugin architecture, and is developed and maintained as an open-source project.

Packmol creates an initial configuration for molecular dynamics simulations by packing molecules into defined regions of space. The packing guarantees that short-range repulsive interactions do not disrupt the simulations. It is developed and maintained as an open-source project.

Atomsk creates, manipulates, and converts atomic systems. It supports many file formats — among them LAMMPS, VASP, Quantum Espresso, IMD, DL_POLY, AtomEye CFG, and xCrySDen XSF — making it easy to convert files for ab initio calculations, classical-potential simulations, or visualization. Atomsk can also perform simple transformations of atomic positions such as rotation, deformation, and inserting dislocations. It is developed and maintained as an open-source project.

polyGraft is developed and maintained by Guang Chen to help build the initial structure and topology of polymer-grafted hybrid systems. A broad range of polymer-grafted hard or soft nanostructures can be generated — polymer-grafted nanoparticles, nanoslabs (planar brushes), nanorods/nanopores, and bottlebrush polymers — accounting for the chain length and grafting density of the polymer graft and the size and shape of the substrate.

Examples of systems built with polyGraft (click for a larger image):

OCTA is an open-source software package consisting of simulation engines (molecular dynamics, rheology simulation, self-consistent field theory, finite element method, and more) and a GUI for visualization, simple molecular building, and analysis of soft-matter systems. It also provides an environment for the collaborative use of several simulators (multi-physics and multi-scale simulations). The commercial J-OCTA adds complex molecular building for full-atomistic and coarse-grained MD.

LAMMPS Interface is a Python program designed as an easy-to-use interface between crystallographic information files (.cif) and LAMMPS. By default it creates LAMMPS input files using the UFF force field from CIF files; the CIF files are expected to be in P1 symmetry.

The Molecular Simulation Design Framework (MoSDeF) is a set of extensible Python tools designed to facilitate the initialization, atom-typing, and screening of soft-matter systems using molecular simulation.

The ATB project is led by Prof. Alan E. Mark (University of Queensland). It provides topology files for organic molecules in formats compatible with LAMMPS and other molecular dynamics packages; the LAMMPS topology files enable building complex systems with the Moltemplate tool distributed with LAMMPS. The site provides topologies for a wide variety of molecules (>20,000 and growing), searchable by name or chemical formula. On a molecule’s page, choose “Molecular Dynamics (MD) Files” and select “LAMMPS” as the output format; molecules not yet in the database can be submitted for processing.

CIF2Cell generates the geometrical setup for various electronic-structure and simulation codes from a CIF (Crystallographic Information Framework) file. It supports output for many programs, including LAMMPS, ABINIT, ASE, CASTEP, CP2K, CPMD, CRYSTAL09, Elk, EMTO, Exciting, Fleur, FHI-aims, MOPAC, Quantum Espresso, RSPt, Siesta, SPR-KKR, and VASP, plus common geometry formats (.coo, .cfg, .xyz). It was published in Computer Physics Communications 182 (2011) 1183–1186, and can be installed via pip or conda-forge.

Particles of specified geometries are typically created by the lattice command in LAMMPS, which can produce rough surfaces when the particle spacing is not small enough — while too small a spacing yields too many particles and higher cost. A similar issue arises when creating atoms from an external STL file. Geoparticle aims to easily construct geometries where smooth surfaces are required. It is hosted on GitHub, documented on Read the Docs, and installable via pip (pip install geoparticle).

JARVIS for force-fields (JARVIS-FF) is a high-throughput computational database for LAMMPS calculations on density-functional-theory-optimized geometric structures with various force fields / interatomic potentials. Its goal is to provide an easy look-up table for evaluating force fields through a web interface and to enhance data reproducibility. JARVIS-FF is part of the Materials Genome Initiative (MGI) at the National Institute of Standards and Technology (NIST).

Hosted by the National Institute of Standards and Technology (NIST), the Interatomic Potentials Repository is a curated source of interatomic potentials (force fields), their parameter files, and evaluation tools. It spans many potential classes (EAM, MEAM, ReaxFF, and others) and material systems, with files submitted or vetted by developers and accompanied by the relevant citations; many are provided in formats ready to use with LAMMPS. The site also publishes computed properties — lattice constants, elastic constants, melting temperatures, and more — so users can compare potentials and pick one suited to their problem.

The Crystallography Open Database (COD) is an open-access collection of crystal structures of organic, inorganic, metal-organic compounds and minerals (excluding biopolymers). It contains over half a million crystal structures in CIF format, which can be converted into LAMMPS input with tools such as cif2cell or LAMMPS Interface.