This notebook downloads results of spectral annotations from classical molecular networking or feature-based molecular networking job from GNPS [http://gnps.ucsd.edu] and generate virtual metabolites either with SyGMa or BioTransformer. The resulting candidates can be used for Network Annotation Propagation on GNPS or with SIRIUS.
Click on the following link to launch the vm-NAP web-app. Note that this is a streamlit temporary instance with limited ressources.
Install locally in conda with:
Download the present repository.
In the terminal, navigate to the repository folder.
Install the environment with:
conda env create --file environment.yml
Initiate the environment:
conda activate vm-NAP
Start the streamlit app:
streamlit run vm-NAP_streamlit.py --server.port 8501 --server.address 0.0.0.0
Representative command for the python script:
python src/vm-NAP_processing.py --job_id='bbee697a63b1400ea585410fafc95723' --run_sygma --run_biotransformer --sirius_input_file 'input/compound_identifications.tsv' --debug --max_compounds_debug=3
Running this for help:
python src/vm-NAP_processing.py --help
The interactive notebook can be accessed via this badge ->
Alternative - The interactive notebook can be accessed via this badge an gesis server->
Note that this is also a temporary instance with limited ressources.
See the documentation for custom database in NAP and how to run Network Annotation Propagation (NAP) on GNPS https://ccms-ucsd.github.io/GNPSDocumentation/nap/#structure-database.
See the documentation to generate the SIRIUS custom database here.
IMPORTANT: Note that only spectral annotations that have a valid InChI or SMILES identifier will be considered downstream. If the annotations you are interested in don't have an identifier in the library, go back to the GNPS library entry, update the entry by adding an identifier, and rerun your GNPS job