AIASSE – Ab Initio Augmented Structure Solving Engine
Overview
AIASSE is a comprehensive software package designed for augmented structure solving by integrating Empirical Potential Structure Refinement (EPSR) with ab initio Density Functional Theory (DFT) calculations. It enables robust, reproducible simulations and analysis for various systems such as simple and complex fluids and fluid mixtures, and networked and non-networked amorphous solids.
Repository Structure
AIASSE/ ├── AIASSE.py - Main Python executable launching the application. ├── Large_box - Scripts and input files for Large Box simulations. ├── Small_box - Scripts and input files for Small Box simulations. ├── Examples - Example inputs, library molecules, and supplementary data. ├── build.sh - Build script to install dependencies and set up the environment. ├── src/ - Internal modules and subroutines used by AIASSE. └── AIASSE_logo/ - Contains the AIASSE.png logo image.
Installation
-
Prerequisites:
Ensure you have Python (version 3.x recommended) and a Unix-like shell (Bash) installed. -
Build the Project:
Open a terminal in the repository root and run:./build.sh
This script installs all required dependencies (including EPSR25 and EPSR26) and sets up the proper directory structure. Upon successful completion, you should see a confirmation message such as:
Build complete! Find your executable in the 'dist' folder.
Usage Workflow
AIASSE follows a multi-step workflow for structure solving:
Initial Setup:
Run the setup module to create the necessary folders (Large_box and Small_box) and copy essential files.
Building Atom/Molecular Geometry:
Use the “Build atom/molecular geometry” tool to either select pre-existing .ato/.mol files from the library or create new ones via the integrated Jmol interface.
Utilize the density converter and Lennard-Jones parameters list if needed.
Running the Large Box Simulation:
Upload your neutron/X-ray scattering data to generate weights.
Create the EPSR input file and start the Large Box simulation.
Monitor simulation parameters (e.g., F(Q), g(r), empirical potential) and extract Reverse Monte Carlo (RMC) configurations once a satisfactory fit is achieved.
Running the Small Box Simulation:
Transfer potential files from the Large Box to the Small Box.
Run the Small Box simulation to generate configurations suitable for DFT analysis (ensure that the atom count is within recommended limits).
DFT Setup and Corrections:
Use provided tools to generate input files for CP2K, CASTEP, etc.
Apply “New Charge” and “New Bond” corrections based on DFT outputs to update charge distributions, bond lengths, bond angles, and dihedral angles.
Iterative Refinement and Material Characterization:
Repeat the simulation and correction cycles until the EPSR fit meets required tolerances.
Analyze bulk properties (e.g., radial distribution functions, coordination numbers) and electronic properties (e.g., density of states, charge distributions) using the post-processing tools.Additional Notes
Some example datasets (e.g., for low density amorphous ice and methanol–water mixtures) are proprietary and are not included in the repository.
A detailed supplementary document (provided in LaTeX) explains the entire workflow with illustrative screenshots.
The repository includes a library of over 900 Lennard-Jones parameter sets for potential selection.
For troubleshooting, refer to the documentation within the Examples folder.License and Commercial Use
Unless explicitly stated otherwise, all material is copyright ©2025 The University of Edinburgh.
Important: The software is not to be used for commercial purposes without the permission of the owners. If you wish to use AIASSE in proprietary, closed-source projects or for other commercial applications, please contact: cip.pruteanu@ed.ac.uk.
Contributors
Ayobami Daramola (adaramo2@ed.ac.uk)
Graeme J. Ackland (gjackland@ed.ac.uk)
Ciprian G. Pruteanu (cip.pruteanu@ed.ac.uk)Contact
For questions, issues, or contributions, please contact the authors at the email addresses listed above.