By combining a Message-Passing Graph Neural Network (MPGNN) and a Forward fully connected Neural Network (FNN) with an integrated gradients explainable artificial intelligence (XAI) method, the authors developed MolGrad and tested it on a number of ADME predictive tasks. MolGrad incorporates explainable features to facilitate interpretation of the predictions. This model has been trained using experimental data on the permeability of molecules across Caco2 cell membranes (Papp, cm s-1)
This model was incorporated on 2021-10-19.
- Ersilia Identifier:
eos1af5 - Slug:
molgrad-caco2
- Task:
Annotation - Subtask:
Activity prediction - Biomedical Area:
ADMET - Target Organism:
Homo sapiens - Tags:
Permeability,ADME,Papp,Chemical graph model
- Input:
Compound - Input Dimension:
1
- Output Dimension:
1 - Output Consistency:
Fixed - Interpretation: Log 10 of the Passive permeability in cm s-1
Below are the Output Columns of the model:
| Name | Type | Direction | Description |
|---|---|---|---|
| log10_passive_permeability | float | high | Log10 of passive permeability |
- Source:
Local - Source Type:
External - DockerHub: https://hub.docker.com/r/ersiliaos/eos1af5
- Docker Architecture:
AMD64 - S3 Storage: https://ersilia-models-zipped.s3.eu-central-1.amazonaws.com/eos1af5.zip
- Model Size (Mb):
17 - Environment Size (Mb):
2418 - Image Size (Mb):
2379.37
Computational Performance (seconds):
- 10 inputs:
34.74 - 100 inputs:
30.01 - 10000 inputs:
815.83
- Source Code: https://github.com/josejimenezluna/molgrad/
- Publication: https://pubs.acs.org/doi/10.1021/acs.jcim.0c01344
- Publication Type:
Peer reviewed - Publication Year:
2021 - Ersilia Contributor: miquelduranfrigola
This package is licensed under a GPL-3.0 license. The model contained within this package is licensed under a AGPL-3.0-only license.
Notice: Ersilia grants access to models as is, directly from the original authors, please refer to the original code repository and/or publication if you use the model in your research.
To use this model locally, you need to have the Ersilia CLI installed. The model can be fetched using the following command:
# fetch model from the Ersilia Model Hub
ersilia fetch eos1af5Then, you can serve, run and close the model as follows:
# serve the model
ersilia serve eos1af5
# generate an example file
ersilia example -n 3 -f my_input.csv
# run the model
ersilia run -i my_input.csv -o my_output.csv
# close the model
ersilia closeThe Ersilia Open Source Initiative is a tech non-profit organization fueling sustainable research in the Global South. Please cite the Ersilia Model Hub if you've found this model to be useful. Always let us know if you experience any issues while trying to run it. If you want to contribute to our mission, consider donating to Ersilia!