Use the latest metabolomics algorithms with a few lines of code
Omigami is an open source Python and R package that gives you access to scalable APIs for the latest metabolomics algorithms
Credit: Courtesy of Richard Yost
Omigami APIs are built in collaboration with the research labs who are at the forefront of novel algorithms as Wageningen, UCDavis, Peter Dorrestein Lab and Oliver Fiehn Lab. Omigami makes it easy to use and test a new algorithm and pre-trained model in your metabolomics pipeline - shortly after publication.
Use the algorithms in less than 5 minutes
Omgami is backed by an auto-scaling kubernetes infrastructure that adjusts to your dataset. Never wait days for a result again.
pip install omigami_client
from omigami_client import Spec2VecClient
client = Spec2VecClient(token="your_token")
mgf_file_path = "path_to_file.mgf"
n_best_matches = 10
result = client.match_spectra_from_path(mgf_file_path,
n_best_matches)
from omigami_client import Spec2VecClient
client = Spec2VecClient(token="your_token")
mgf_file_path = "path_to_file.mgf"
n_best_matches = 10
result = client.match_spectra_from_path(mgf_file_path,
n_best_matches)
Long term support
Omigami APIs are maintained, documented and supported by a full-time team of researchers and machine learning engineers - and funded by a auto-renewing grant from Data Revenue.
APIs that scale with your dataset
Omgami is backed by an auto-scaling kubernetes infrastructure that adjusts to your dataset. Never wait days for a result again.
Free forever
Omigami is free for academic use. Financing comes exclusively from commercial projects. If you want to use Omigami APIs for a commercial project reach out to us
Transparent algorithms
The algorithms behind all APIs are open source. You can find detailed information on each and every model in the documentation
Try a better algorithms - upgrade your pipeline today
MS2DeepScore
Supervised deep learning model trained to predict structural similarity directly from MS-spectra.
MS2DeepScore
Supervised deep learning model trained to predict structural similarity directly from MS-spectra.
Coming soon