This is a project to leverage Deep learning to predict drug-drug interactions.
Drug Interactions are a major cause for drug recalls and affect 5% to 30 % of all adverse events.
Further details regarding the motivation, methods and results of implementing this idea can be found in my presentation here : http://bit.ly/DrugInteractionSlides
The package as well as the necessary requirements can be installed by running make or via
python setup.py install
For Training the model run the following train.sh script - there are 3 alternatives, ECFP, SMILES, Transfer_Learning
For Inference with multiple drugs, use the inference_from_file.sh script. For single inference use the inference.sh script