LipiDetective is a deep learning framework designed for the identification of molecular lipid species from tandem mass spectra.
- Deep Learning Framework: Utilizes the transformer architecture to identify lipid species in shorthand nomenclature format.
- Configurable Input: Accepts YAML configuration file for streamlined and reproducible setup.
- Data Handling: Supports training with HDF5 files and prediction using mzML files.
The easiest way to get started with LipiDetective is using Poetry and the provided pyproject.toml file. Make sure you have at least python version 3.11 installed. Follow these steps to set up the project:
- Clone the repository:
git clone https://github.com/LipiTUM/lipidetective.git cd lipidetective - Install the dependencies:
poetry install
- If necessary activate the virtual environment:
poetry shell
LipiDetective can be executed by running the following command:
poetry run lipidetective --config path/to/config.yamlEnsure that the YAML configuration file is correctly set up with the necessary parameters for the desired workflow.
The configuration file should be in YAML format and contain paths to your HDF5 training files or mzML prediction files,
along with any additional parameters required for the model. A template is provided in the folder config/config_templates.
If you want to use LipiDetective for your own work, please cite our manuscript
@article{Wuerf:2024,
title={LipiDetective - a deep learning model for the identification of molecular lipid species in tandem mass spectra},
author={W\"urf, Vivian and K\"ohler, Nikolai and Molnar, Florian and Hahnefeld, Lisa and Gurke, Robert and Witting, Michael and Pauling, Josch K},
journal={bioRxiv},
year={2024},
publisher={Cold Spring Harbor Laboratory},
doi={https://doi.org/10.1101/2024.10.07.617094}
}
This project is licensed under the BSD 3-Clause License - see the LICENSE file for details.