It is also possible to run DreaMS remotely in Google Colab, as demonstrated in the [tutorial](https://drive.google.com/file/d/1ZUGsci4UCq-92fBgONKt6D_Nxo1uFw1-/view?usp=sharing) for the DreaMS-based chemical profiling of beer samples. This tutorial shows how to:
1. Extract and preprocess MS/MS spectra from LC-MS/MS experiments.
2. Compare beers by analyzing molecular embeddings (MS/MS spectrum embeddings).
3. Build a taxonomic tree of beers by analyzing experiment-level embeddings (LC-MS/MS experiment embeddings).
## Getting started locally
### Installation
Run the following code from the command line.
``` shell
# Download this repository
git clone https://github.com/pluskal-lab/DreaMS.git
cd DreaMS
# Create conda environment
conda create -n dreams python==3.11.0 --yes
conda activate dreams
# Install DreaMS
pip install -e .
```
If you are not familiar with conda or do not have it installed, please refer to the [official documentation](https://conda.io/projects/conda/en/latest/user-guide/getting-started.html).
### Compute DreaMS representations
To compute DreaMS representations for MS/MS spectra from `.mgf` file, run the following Python code.
``` python
from dreams.api import dreams_embeddings
embs = dreams_embeddings('data/examples/example_5_spectra.mgf')
```
The resulting `embs` object is a matrix with 5 rows and 1024 columns, representing 5 1024-dimensional DreaMS representations for 5 input spectra stored in the `.mgf` file.
## References
- Paper: [https://www.nature.com/articles/s41587-025-02663-3](https://www.nature.com/articles/s41587-025-02663-3).
- Pre-print: [https://chemrxiv.org/engage/chemrxiv/article-details/6626775021291e5d1d61967f](https://chemrxiv.org/engage/chemrxiv/article-details/6626775021291e5d1d61967f).
- Web app on Hugging Face Spaces: [https://huggingface.co/spaces/anton-bushuiev/DreaMS](https://huggingface.co/spaces/anton-bushuiev/DreaMS).
- Documentation and tutorials: [https://dreams-docs.readthedocs.io/](https://dreams-docs.readthedocs.io/).
- Weights of pre-trained models: [https://zenodo.org/records/10997887](https://zenodo.org/records/10997887).
- Datasets:
- GeMS dataset: [https://huggingface.co/datasets/roman-bushuiev/GeMS/tree/main/data](https://huggingface.co/datasets/roman-bushuiev/GeMS/tree/main/data).
- DreaMS Atlas: [https://huggingface.co/datasets/roman-bushuiev/GeMS/tree/main/data/DreaMS_Atlas](https://huggingface.co/datasets/roman-bushuiev/GeMS/tree/main/data/DreaMS_Atlas).
- Labeled MS/MS spectra: [https://huggingface.co/datasets/roman-bushuiev/GeMS/tree/main/data/auxiliary](https://huggingface.co/datasets/roman-bushuiev/GeMS/tree/main/data/auxiliary).
If you use DreaMS in your research, please cite the following paper:
```bibtex
@article{bushuiev2025selfsupervised,
author={Bushuiev, Roman
and Bushuiev, Anton
and Samusevich, Raman
and Brungs, Corinna
and Sivic, Josef
and Pluskal, Tom{\'a}{\v{s}}},
title={Self-supervised learning of molecular representations from millions of tandem mass spectra using DreaMS},
journal={Nature Biotechnology},
year={2025},
month={May},
day={23},
abstract={Characterizing biological and environmental samples at a molecular level primarily uses tandem mass spectroscopy (MS/MS), yet the interpretation of tandem mass spectra from untargeted metabolomics experiments remains a challenge. Existing computational methods for predictions from mass spectra rely on limited spectral libraries and on hard-coded human expertise. Here we introduce a transformer-based neural network pre-trained in a self-supervised way on millions of unannotated tandem mass spectra from our GNPS Experimental Mass Spectra (GeMS) dataset mined from the MassIVE GNPS repository. We show that pre-training our model to predict masked spectral peaks and chromatographic retention orders leads to the emergence of rich representations of molecular structures, which we named Deep Representations Empowering the Annotation of Mass Spectra (DreaMS). Further fine-tuning the neural network yields state-of-the-art performance across a variety of tasks. We make our new dataset and model available to the community and release the DreaMS Atlas---a molecular network of 201 million MS/MS spectra constructed using DreaMS annotations.},
issn={1546-1696},
doi={10.1038/s41587-025-02663-3},
url={https://doi.org/10.1038/s41587-025-02663-3}
}
```