AutoLEI: An XDS-based pipeline with graphical user interface for automated real-time and offline batch 3D ED/microED data processing

AutoLEI is an XDS-based pipeline with graphical user interface for automated real-time and offline batch 3D ED/microED data processing. It provides a user-friendly platform for rapid, automated data processing and merging of multiple MicroED datasets, with well-designed and significantly streamlined structure determination workflows.

If you use AutoLEI in your work, please cite the following publication:

1. Wang, L., Chen, Y., Hutchinson, E. S., Stenmark, P., Hofer, G., Xu, H., & Zou, X. (2026). IUCrJ, 13. https://doi.org/10.1107/S2052252525010784 

2. Kabsch, W. (2010). Acta Crystallogr D Biol Crystallogr 66, 125–132. https://doi.org/10.1107/S0907444909047337

If you encounter any issues, please report them to:
lei.wang@su.se
yinlin.chen@su.se

Version selection:

AutoLEI v-1.0.3.260305 (Latest build)

AutoLEI v-1.0.0: version demonstrated in the paper (https://doi.org/10.1101/2025.04.12.648515).

Data available:

https://zenodo.org/records/14536385 

What's new in v-1.0.3:

- SHELX INS file generation is now supported when SHELX HKL is generated via XDSCONV.
- The Laue group recognition algorithm has been updated, delivering a higher success rate and improved accuracy.
- The phase recognition algorithm has also been improved.
- With AutoSolveX enabled, real-time mode can now identify the Laue group and solve the structure automatically, provided the corresponding setting is activated.
- A new framework has been reconstructed in preparation for the upcoming version 1.1 update.
- The rotation angle refinement algorithm has been rewritten, achieving 3× faster performance.
- HKL analysis has also been accelerated by 3×.
- The slice viewer and 3D reciprocal space viewer now feature a new GUI.
- Numerous bugs have been fixed.

If you want to update autolei, in the Python environment type: pip install -U autolei

1. Key Features
User-friendly interface: Simplifies MicroED data processing, requiring minimal manual input.
-Batch Processing: Handles large numbers of datasets with automated workflows.
-Real-Time Data Processing: Provides live feedback during data collection.
Versatility: Supports diverse samples, including small molecules and protein workflows.

2. Installation

Operating Systems: Linux or Windows via [WSL](https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux) (versions 1/2).
Software Dependencies:
Python 3.8+ with libraries specified in `pyproject.toml`.
[XDS](https://xds.mr.mpg.de/) and [XDSGUI](https://wiki.uni-konstanz.de/xds/index.php/XDSGUI).
Optional tools: `xprep` for advanced features and LibreOffice for `.xlsx` files in Linux.

Steps:
a) Install via pip:
   bash
   pip install autolei

   For historical versions, use:
   bash
   pip install autolei-[version_name].zip
  
b) Manual installation:
   Follow the steps in the AutoLEI_Tutorial.pdf and https://gitlab.com/tristonewang/autolei/-/wikis/home 

3.  Usage
Command-line Usage

Launch the GUI
   bash
   autolei

Note: The first launch may take slightly longer as dependencies initialize.

Configure Settings:
   bash
   autolei_setting
   
  The opened .ini file includes settings on screen scaling, multi-thread and report format.

Import Instrument:
   bash
   autolei_add_instrument [instrument_setting_file]

4. GUI pages
AutoLEI is organised into multiple working pages:
- Input: Configure experiment parameters and generate input files.
- XDSRunner: Automate initial processing and data quality inspection.
- CellCorr: Update unit cell information and refine settings.
- XDSRefine: Fine-tune processing parameters, including rotation axis and scaling.
- MergeData: Filter and merge datasets for downstream analysis.
- Cluster&Output: Perform clustering and generate outputs for structure determination.
- Expert: Miscellaneous tools for data reduction and PETS2-related function.
- RealTime: Live data processing with real-time feedback and automatic merging.

5. Documentation and other sources
Detailed guides and examples can be found in:

- Tutorial for AutoLEI
- PyPI: https://pypi.org/project/autolei/ 
- Gitlab (our Wiki): https://gitlab.com/tristonewang/autolei/-/wikis/home

6. Authors and Acknowledgments
Developed by Lei Wang and Yinlin Chen. Contributions from Gerhard Hofer, Hongyi Xu, and Xiaodong Zou at Stockholm University. The project integrates valuable resources from [edtools](https://github.com/instamatic-dev/edtools).

7. License
The software is licensed under the BSD 3-Clause License.