wilhelm-lab/koina
Democratizing ML in proteomics
This tool helps proteomics researchers, lab technicians, and mass spectrometry specialists predict the outcome of fragmentation experiments. You input peptide sequences, their collision energies, and precursor charges, and it outputs predicted fragment ion annotations, their mass-to-charge ratios (m/z), and their intensities. This allows for more accurate identification and quantification of peptides.
Use this if you need to predict mass spectrometry fragmentation patterns for peptides to support protein identification or quantification workflows.
Not ideal if your work doesn't involve proteomics, mass spectrometry, or peptide analysis.
Stars
52
Forks
22
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Mar 10, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/wilhelm-lab/koina"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
CompOmics/DeepLC
DeepLC: Retention time prediction for peptides carrying any modification.
BojarLab/glycowork
Package for processing and analyzing glycans and their role in biology.
wilhelm-lab/dlomix
Python framework for Deep Learning in Proteomics
BojarLab/CandyCrunch
Predicting glycan structure from LC-MS/MS data
sidhomj/DeepTCR
Deep Learning Methods for Parsing T-Cell Receptor Sequencing (TCRSeq) Data