Lightning Talk

AutoMapper: A Tool for Accelerating the Fix Bond/React Workflow

Matthew A. Bone
University of Bristol
  • Thursday, 12 Aug 2021
  • 11:00 - 11:03 EDT
  • Prerecorded Video

Fix bond/react is a powerful function within LAMMPS that enables dynamic bonding of covalent polymers. This is much improved over the traditional stop-start approach to complex polymer bonding schemes seen in the literature [1,2]. However, pre-processing of the two required “pre-bond” and “post-bond” molecule templates, and associated map file, is a time-consuming and challenging part of the LAMMPS workflow. AutoMapper aims to eliminate this pre-processing effort by automatically converting the pre-bond and post-bond LAMMPS input file to molecule templates and generates a map file with minimal input from the user. This considerably reduces the manual data handling and can be coupled other third-party tools, such as Moltemplate, which automate the generation of LAMMPS input files. AutoMapper is a key step in automating the LAMMPS pre-processing stage and will help to facilitate high throughput MD simulations for data analysis and machine learning.

The mapping component of AutoMapper uses an adapted path search method that relies on basic chemical intuition to correctly map pre-bond atom IDs to their post-bond equivalents. By comparing the neighbours of known mapped atoms, AutoMapper discovers new atom pairs which propagate the search forward. Utilising a modest atomic fingerprint system and missing atom buffers enables the path search to successfully map more complex molecules with symmetry and structural rearrangement on bonding. AutoMapper has been proven accurate on a range of different test cases and will be released publicly on GitHub for wider use.


  1. Bone, M.A.; Macquart, T.; Hamerton, I.; Howlin, B.J. A Novel Approach to Atomistic Molecular Dynamics Simulation of Phenolic Resins Using Symthons. Polymers 2020, 12, 926.

  2. Demir, B.; Walsh, T.R. A robust and reproducible procedure for cross-linking thermoset polymers using molecular simulation. Soft Matter 2016, 12, 2453–2464.