This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| GROMOS | |
|---|---|
| Name | GROMOS |
| Title | GROMOS |
| Developer | University of Groningen; University of Zürich; ETH Zurich |
| Released | 1970s |
| Latest release | GROMOS 54A8 / 2016* |
| Programming language | Fortran (programming language), C (programming language) |
| Operating system | Linux, Unix, Windows |
| License | Academic/Research |
GROMOS is a family of empirical force fields and an associated molecular dynamics simulation engine originally developed for biomolecular and condensed-phase simulations. It was established by researchers at University of Groningen and further advanced at ETH Zurich and University of Zürich to model proteins, nucleic acids, carbohydrates, lipids, and small molecules in condensed phases. The project influenced computational studies at institutions such as Max Planck Society, European Molecular Biology Laboratory, and industrial groups at Roche and Novartis.
The project traces roots to work in the 1970s at University of Groningen and early collaborations with University of Basel researchers, evolving through major contributions in the 1980s and 1990s at ETH Zurich and University of Zürich. Key milestones include parameter sets developed during the era of researchers affiliated with Royal Society fellowships and funding from the European Research Council. GROMOS parameters were refined alongside advances in hardware at centers like CERN and supercomputing facilities such as Oak Ridge National Laboratory and Lawrence Livermore National Laboratory. Over decades the code and parameter sets passed through contributions by groups connected to Swiss National Science Foundation, Deutsche Forschungsgemeinschaft, and industrial collaborations with Pfizer and Merck & Co..
GROMOS is an empirical, united-atom force field family using bonded and nonbonded terms parameterized to reproduce experimental thermodynamic properties measured at institutions like National Institute of Standards and Technology and spectroscopic observables from groups at Max Planck Institute for Biophysical Chemistry. Parameterization strategies referenced experimental data from International Union of Pure and Applied Chemistry datasets and quantum calculations associated with research at California Institute of Technology and Massachusetts Institute of Technology. Major variants include parameter sets tuned for condensed-phase free energies and solvation, with optimization methods influenced by algorithms developed at Royal Society laureate groups and numerical techniques from Los Alamos National Laboratory. GROMOS employs Lennard-Jones potentials and tailored dihedral terms calibrated against high-level ab initio benchmarks from ETH Zurich collaborators.
The reference implementation originated as a Fortran codebase maintained by groups at University of Groningen and later by maintainers at University of Zürich. Parallel and optimized builds have been produced to run on clusters at Argonne National Laboratory and accelerators used by IBM systems. Related toolchains for topology construction and analysis integrate with software from European Bioinformatics Institute, Protein Data Bank, Amber utilities, and conversion scripts used in GROMACS pipelines. Community contributions include ports and wrappers that interface with packages developed at University of Cambridge and visualization support compatible with tools from RCSB PDB and UCSF.
GROMOS parameter sets and code have been applied to simulate protein folding studies influenced by questions tackled at Cold Spring Harbor Laboratory and ligand binding investigations relevant to projects at GlaxoSmithKline and AstraZeneca. Simulations of membrane systems drew comparisons with experiments from Max Planck Institute for Colloids and Interfaces and techniques used at Lawrence Berkeley National Laboratory beamlines. Thermodynamic integration and free-energy perturbation calculations using GROMOS informed drug-design workflows at Novartis and academic collaborations with University of Oxford and University of Cambridge. Studies of carbohydrate dynamics engaged researchers affiliated with Karolinska Institutet and University of Copenhagen.
Validation efforts benchmarked GROMOS against experimental observables reported by National Institutes of Health, neutron scattering results from Institut Laue–Langevin, and calorimetry datasets produced at European Molecular Biology Laboratory. Comparative tests used standards and protocols similar to those developed at MolSSI and computational challenges organized by CASP and D3R Grand Challenge participants. Performance and accuracy benchmarks were run on infrastructures at PRACE and national supercomputing centers, with metrics including density, heat of vaporization, and solvation free energies measured relative to data from NIST Chemistry WebBook archives.
GROMOS has been compared extensively to popular alternatives such as AMBER (force field), CHARMM, OPLS-AA, and Martini (force field). Differences center on united-atom versus all-atom representations, parametrization targets, and treatment of polarizability—issues debated in workshops at EMBO and methodological conferences at Gordon Research Conferences. Studies contrasting GROMOS with AMBER and CHARMM used benchmark systems common to research groups at University of California, San Diego and Yale University, revealing trade-offs in reproducing condensed-phase thermodynamics versus gas-phase conformational energetics.
Development is coordinated by academic consortia with contributors from University of Groningen, ETH Zurich, University of Zürich, and partner labs at Max Planck Society and European Molecular Biology Laboratory. Community support and training occur via workshops at EMBO, tutorials at Gordon Research Conferences, and summer schools hosted by University of Cambridge and University of Oxford. Collaborative repositories and user forums include academic mirrors and project pages maintained at institutional servers associated with Swiss National Supercomputing Centre and national consortia. Ongoing work focuses on expanding parameters for drug-like molecules in collaboration with industrial partners such as Roche and BASF.