What is it about?
Molecular dynamics simulations of ice formation produce trajectories with millions of atomic positions per frame. Determining what type of ice (or liquid) each molecule belongs to requires structural analysis algorithms that examine local bonding patterns. These algorithms existed in the literature but lacked accessible, maintained implementations. d-SEAMS provides open-source implementations of state-of-the-art structural analysis methods for ice and water. The code reads standard trajectory formats (LAMMPS, GROMACS), identifies ice polymorphs (Ih, Ic, clathrates) using topological network criteria, and outputs per-frame structural assignments suitable for further analysis. We demonstrated d-SEAMS on simulations of water freezing in carbon nanotubes, where exotic ice phases form under confinement. The code identified transitions between phases that would have been difficult to detect by visual inspection.
Featured Image
Photo by Megs Harrison on Unsplash
Why is it important?
Structural analysis is a bottleneck in simulation studies of ice formation, clathrate nucleation, and water under confinement. Without automated tools, researchers resort to visual inspection of snapshots, which does not scale and misses transient structures. d-SEAMS makes published structural analysis algorithms available as tested, documented software. The topological criteria implemented (based on ring statistics and bond-order parameters) are the same ones used in the original research papers, so results are directly comparable to the literature. The code handles the large trajectory files produced by modern simulations (millions of atoms, thousands of frames) efficiently, using Lua scripting for user-defined analysis pipelines.
Perspectives
This was my first co-lead author paper, with Amrita Goswami, who developed the structural analysis algorithms and implemented them. My contribution was the software engineering: the build system, I/O layer, testing framework, and documentation. The project taught me that the gap between a working research code and a usable software package is larger than most researchers expect. Getting d-SEAMS to install cleanly on different systems, handle edge cases in trajectory files, and produce reproducible results across platforms took as much effort as the algorithm implementation. The code is still maintained and used for ongoing ice nucleation research.
Rohit Goswami
University of Iceland
Read the Original
This page is a summary of: d-SEAMS: Deferred Structural Elucidation Analysis for Molecular Simulations, Journal of Chemical Information and Modeling, March 2020, American Chemical Society (ACS),
DOI: 10.1021/acs.jcim.0c00031.
You can read the full text:
Resources
Contributors
The following have contributed to this page
