SteroidXtract is a deep-learning based bioinformatics tool for steroid MS2 extraction from untargeted metabolomics data.
SteroidXtract is written in Python and is publicly available for non-commercial use.
In "toolkits" folder, data augmentation methods and nested cross-validation are provided.
For detailed instructions, please refer to the user manual of SteroidXtract.
Update (Oct 10, 2024): The original version of SteroidXtract works with the following software versions
- Python 3.8.0 (in virtual enviroment using
virtualenv) - Pandas 2.0.3
- Numpy 1.22 (downgraded from newest version, after pandas installation)
- pyteomics 4.7.4
- lxml 5.3.0
- Tensorflow (as
tensorflow-cpu) 2.13.0
To download the SteroidXtract CNN model, please go to the Supporting Information (zip file) of our paper. https://pubs.acs.org/doi/abs/10.1021/acs.analchem.0c04834
S. Xing, Y. Jiao, M. Salehzadeh, K. Soma, T. Huan, ‘SteroidXtract: Deep Learning-Based Pattern Recognition Enables Comprehensive and Rapid Extraction of Steroid-Like Metabolic Features for Automated Biology-Driven Metabolomics’, Analytical Chemistry, 2021(93), 5735−5743.