CIF Cleaner

CIF Cleaner is an interactive and codeless program for sorting, pre-processing, and parsing CIF (Crystallographic Information File) files. It extracts information from .cif files in a folder containing such files.

CIF Cleaner is built using the cifkit package: https://github.com/bobleesj/cifkit.

Motivation

In high-throughput analysis, we extract .cif files from databases, sorting them based on specific attributes. CIF Cleaner was developed to streamline this process by utilizing interactive prompts.

This tool is used by undergraduate and graduate researchers in the Oliynyk research group at CUNY Hunter College.

Demo

Getting started

Open your terminal or command-line interface:

1. Clone the repository and enter the directory:

git clone https://github.com/bobleesj/cif-cleaner.git
cd cif-cleaner

2. Install dependencies from the requirements.txt file:

pip install -r requirements.txt

Run the program:

python main.py

The program automatically detects folders containing .cif files at the project level.

Welcome! Please choose an option to proceed:
[1] Move files based on unsupported format after pre-formatting
[2] Move files based on unreasonable distance
[3] Move files based on supercell atom count
[4] Move files based on tags
[5] Move files based on composition type
[6] Move files based on elements
[7] Move files based on coordination number
[8] Copy files based on atomic occupancy and mixing
[9] Get file info in the folder

Enter your choice (1-9): 6
You have chosen: Get file info in the folder

Available folders containing CIF files:
1. elements CIF, 1787 files
2. binary R-X CIF, 3757 files
3. binary R-M CIF, 5510 files
4. binary M-X CIF, 2026 files
5. 20240706_test, 28 files
6. ternary R-M-X CIF, 11418 files

Enter the number corresponding to the folder containing .cif files: 1

Options

Option	Description	User input
1	Standardize and relocate .cif files that do not comply with the expected format	-
2	Move .cif by minimum distance threshold	Min distance (e.g., 2.0 Å)
3	Move .cif by supercell size	Min and max atom count
4	Move .cif by tags	-
5	Move .cif by composition type i.g. binary, ternary, etc.	-
6	Move .cif by input elements, matching or containing	Elements (e.g., Er Co)
7	Move .cif by input coordination numbers, matching or containing	Coordination numbers (e.g., 12 16)
8	Copy .cif by atomic mixing, i.g. full occupancy, atomic mixing, etc.	-
9	Get information from .cif files and save .csv	-

Option 2: Filter files by minimum distance

For Option [2], the following histogram is saved in the folder. You need to enter the threshold distance. Any files below this threshold will be moved to a separate folder.

Enter the threshold distance (unit in Å): 2

A folder called min_dist_below_2.0 is created, and files with the minimum atomic distance below 2.0 Å are moved.

Option 3. Filter by supercell size

A supercell is generated by applying a ±1 shift from the unit cell

The program prompts you to enter the minimum and maximum supercell atom count:

Enter the min number of atoms in the supercell: 300
Enter the max number of atoms in the supercell: 500

A folder called supercell_above_300_below_500 will be created with files that meet the criteria.

Option 9. Info

A .csv is generated containing containing information for each .cif file in the selected folder.

Filename	Formula	Structure	Tag	Supercell atom count	Site mixing type	Composition type	Min distance (Å)	Processing time (s)
300237	LaPtSb	CaIn2		423	full_occupancy_atomic_mixing	3	2.727	0.301
301710	YFeSi3	Ce(Ni0.59Sb0.41)4		558	full_occupancy_atomic_mixing	3	2.293	0.329
1955459	PrNi0.48Si1.48	AlB2		267	deficiency_atomic_mixing	3	2.328	0.181
1949113	Ce2.11Pd1.92In0.89	Mo2FeB2		437	deficiency_atomic_mixing	3	2.431	0.254
301188	Yb0.5Co3Ge3	Y0.5Co3Ge3		323	deficiency_without_atomic_mixing	3	1.2	0.183
1942974	Dy0.0011Fe0.8291Ga0.1698	W		162	full_occupancy_atomic_mixing	3	2.514	0.032
1940621	LuPd0.5Si1.5	AlB2		267	deficiency_atomic_mixing	3	2.325	0.186
1200981	Ni3Sn2	Ni3Sn2	rt	594	full_occupancy	2	2.503	0.317
301180	Lu0.5Co3Ge3	Y0.5Co3Ge3		323	deficiency_without_atomic_mixing	3	1.197	0.187

Other tools

In addition to CIF Cleaner, there are other interactive tools available that allow you to extract geometric features and visualize bonding patterns from .cif files. I invite you to explore these tools to enhance your analysis:

• Cif Bond Analyzer (CBA): conduct bonding, coordination, and site analysis. https://github.com/bobleesj/structure-analyzer-featurizer
• Structure Analysis/Featurizer (SAF): extract ML features for binary and ternary compounds. https://github.com/bobleesj/cif-bond-analyzer

How to contribute

• the repository and recommend it to your colleagues who might find this repository helpful.
• the repository and request a pull request for any changes.
• If you have any questions or have any suggestions, feel free to reach out to Sangjoon Bob Lee (@bobleesj) for any questions.

Changelog

• 20240709: implement cifki v1.0.2, preprocess for option 1 only (#24)
• 20240708: integreated CI tests for all options
• 20240705: integrated cikift, refactored code and prompts
• 20240303: added GitHub CI integration