This project has been archived by its maintainers, and is no longer receiving any updates.
ms-toolkit
Tools for mass spectrometry (MS) library searching and model training.
This library provides a pipeline for vectorizing spectra, training Word2Vec models, preselecting candidates using clustering/GMM, and searching using weighted cosine or embedding similarity. Portions of the code are adapted from the Spec2Vec project.
Features
- Parse MS library text files with optional progress UI
- Create
SpectrumDocumentobjects for Word2Vec training - Train and load Word2Vec models (
w2v.py) - Vectorize spectra and perform similarity search (
preprocessing.py,similarity.py) - Preselect candidates using KMeans or Gaussian Mixture Models (
preselector.py) - High-level
MSToolkitfacade wrapping the workflow (api.py)
Installation
Install with pip using the included setup.py:
pip install .
Dependencies include numpy, joblib, gensim, and scikit-learn. Optional UI
features require customtkinter or PySide6.
Quick Start: Open MassBank Workflow
Pretrained models for MassBank are provided for immediate use.
Note: MassBank is an open source mass spectral library for small molecule identification. It is freely available and can be downloaded directly through the ms-toolkit API.
from ms_toolkit.api import MSToolkit
# Initialize toolkit (defaults to open MassBank workflow)
toolkit = MSToolkit()
# Download and load the MassBank library (first run will download and cache)
toolkit.download_library()
# Load pretrained MassBank Word2Vec and preselector models
toolkit.load_w2v('models/massbank_25epochs.model')
toolkit.load_preselector('models/massbank_kmeans.pkl')
# Search using a query spectrum (list of (m/z, intensity) tuples)
query = [
(27.0, 0.09),
(39.0, 0.04),
(41.0, 0.09),
(43.0, 0.33),
(71.0, 0.28),
(114.0, 0.05),
]
results = toolkit.search_w2v(query)
for compound, score in results:
print(f'{compound}: {score:.3f}')
Advanced Usage
You can also train your own models or use other libraries (e.g., NIST) if available:
# Example: Train your own Word2Vec model
toolkit = MSToolkit(library_txt="your_library.txt")
toolkit.load_library()
toolkit.vectorize_library()
toolkit.train_preselector(save_path="my_kmeans.pkl")
toolkit.train_w2v(save_path="my_w2v.model")
License
This project is licensed under the Apache License 2.0. See LICENSE for details.
The NOTICE file explains that some code derives from Spec2Vec, which is also
Apache 2.0 licensed.
Metadata
Release files for ms-toolkit-nrel 0.1.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ms_toolkit_nrel-0.1.2.tar.gz | 27.3 kB | With provenance | Details for ms_toolkit_nrel-0.1.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ms_toolkit_nrel-0.1.2-py3-none-any.whl | Python 3 | none | any | Details for ms_toolkit_nrel-0.1.2-py3-none-any.whl |
Total release size: 56.0 kB
Release files / ms_toolkit_nrel-0.1.2.tar.gz
| Download URL | ms_toolkit_nrel-0.1.2.tar.gz |
|---|---|
| Size | 27.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
c28ea88528c8a96a753ca3b946038b6db28722c1c03fe76e960745b90c509930
|
|
BLAKE2b-256 checksum How to use checksums |
a2852bb535799612fc24679d11d095e0fa564b00575de744322f17254954671f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.12.9
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jun 16, 2025.
Transparency logRelease files / ms_toolkit_nrel-0.1.2-py3-none-any.whl
| Download URL | ms_toolkit_nrel-0.1.2-py3-none-any.whl |
|---|---|
| Size | 28.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4bf73666464cc77eb7b8a7ec02c1294508054fb6c122eb9d7220f8abe25f505c
|
|
BLAKE2b-256 checksum How to use checksums |
5e4e48a872e927fbd46a782cbec705ce921be53320713c875113f37bb21f0d22
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.12.9
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jun 16, 2025.
Transparency logSecurity for ms-toolkit-nrel 0.1.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for ms_toolkit_nrel-0.1.2.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jun 16, 2025.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.