slices 1.4.12

Invertible crystal representation (SLICES)

Simplified Line-Input Crystal-Encoding System

This software implementes Simplified Line-Input Crystal-Encoding System (SLICES), the first invertible and invariant crystal representation.

It has several main functionalities:

• Encode crystal structures into SLICES strings
• Reconstruct original crystal structures from their SLICES strings (Text2Crystal)
• Generate crystals with desired properties using conditional RNN (Inverse Design)

Developed by Hang Xiao 2023.04 [e-mail]

[Paper] [Data/Results] [Source code]

We also provide a codeocean capsule (a modular container for the software environment along with code and data, that runs in a browser), allowing one-click access to guaranteed computational reproducibility of SLICES's benchmark. [Codeocean Capsule]

Table of Contents

• Installation
• Examples
  • Crystal to SLICES and SLICES to crystal
  • Augment SLICES and canonicalize SLICES
• Documentation
• Reproduction of benchmarks and inverse design case study
  • General setup
  • Reconstruction benchmark for MP-20
  • Reconstruction benchmark for MP-21-40
  • Reconstruction benchmark for QMOF-21-40
  • Inverse design case study
  • Material generation benchmark
  • Property optimization benchmark
• Citation
• Contact

Installation

Prerequisites: m3gnet, pygraphviz, graphviz, smact. Tip: using pip or conda or apt install/yum install to install these packages.

pip install slices

Examples

Crystal to SLICES and SLICES to crystal

Converting a crystal structure to its SLICES string and converting this SLICES string back to its original crystal structure. Suppose we wish to convert the crystal structure of NdSiRu (mp-5239,https://next-gen.materialsproject.org/materials/mp-5239?material_ids=mp-5239) to its SLICES string and converting this SLICES string back to its original crystal structure. The python code below accomplishes this:

import os
from invcryrep.invcryrep import InvCryRep
from pymatgen.core.structure import Structure
# setup modified XTB's path
# obtaining the pymatgen Structure instance of NdSiRu
original_structure = Structure.from_file(filename='NdSiRu.cif')
# creating an instance of the InvCryRep Class (initialization)
backend=InvCryRep()
# converting a crystal structure to its SLICES string
slices_NdSiRu=backend.structure2SLICES(original_structure) 
# converting a SLICES string back to its original crystal structure and obtaining its M3GNet_IAP-predicted energy_per_atom
reconstructed_structure,final_energy_per_atom_IAP = backend.SLICES2structure(slices_NdSiRu)
print('SLICES string of NdSiRu is: ',slices_NdSiRu)
print('\nReconstructed_structure is: ',reconstructed_structure)
print('\nfinal_energy_per_atom_IAP is: ',final_energy_per_atom_IAP,' eV/atom')
# if final_energy_per_atom_IAP is 0, it means the M3GNet_IAP refinement failed, and the reconstructed_structure is the ZL*-optimized structure.

Augment SLICES and canonicalize SLICES

Converting a crystal structure to its SLICES string and perform data augmentation (2000x), then reduce these 2000 SLICES to 1 canonical SLICES with get_canonical_SLICES.

import os
from invcryrep.invcryrep import InvCryRep
from pymatgen.core.structure import Structure
from pymatgen.analysis.structure_matcher import StructureMatcher, ElementComparator
# obtaining the pymatgen Structure instance of Sr3Ru2O7
original_structure = Structure.from_file(filename='Sr3Ru2O7.cif')
# creating an instance of the InvCryRep Class (initialization)
backend=InvCryRep(graph_method='econnn')
# converting a crystal structure to its SLICES string and perform data augmentation (2000x)
slices_list=backend.structure2SLICESAug(structure=original_structure,num=2000) 
slices_list_unique=list(set(slices_list))
cannon_slices_list=[]
for i in slices_list_unique:
    cannon_slices_list.append(backend.get_canonical_SLICES(i))
# test get_canonical_SLICES
print(len(slices_list),len(set(cannon_slices_list)))
# 2000 SLICES generated by data augmentation has been reduced to 1 canonical SLICES

Documentation

The SLICES documentation is hosted at read-the-docs.

Citation

@article{xiao2023invertible, title={An invertible, invariant crystallographic representation for inverse design of solid-state materials using generative deep learning}, author={Xiao, Hang and Li, Rong and Shi, Xiaoyang and Chen, Yan and Zhu, Liangliang and Wang, Lei and Chen, Xi}, year={2023} }

Contact and Support

Start a new discussion thread in [Discussion], or reach out to Hang Xiao (https://www.researchgate.net/profile/Hang-Xiao-8) [e-mail] if you have any questions.