Project Rephetio Browser

Drug repurposing predictions

Project Rephetio

Project Rephetio is an open science project to predict new uses for existing compounds, which is called drug repurposing. Predictions are created from Hetionet v1.0, an integrative network of biomedicine that contains 2,250,197 relationships of 24 types.

We use machine learning to systematically learn network patterns of drug efficacy. Our method translates the network paths between a compound and disease into a predicted probability of treatment. We make predictions for 1,538 approved small molecule compounds and 136 complex diseases, resulting in a total of 209,168 compound–disease pairs.


Navigate to your compound or disease of interest to see all of its predictions. Treatments refers to the number of disease-modifying therapies for a compound or disease. AUROC measures the ability of the predictions to prioritize these known treatments. Edges refers to the number of relationships in Hetionet v1.0 that connect the compound or disease. In general, the more edges, the more Hetionet knows about a node.

Generating Table...

More information

For more information on Project Rephetio or for feedback, please see the project manuscript and Thinklab page.

Project Rephetio predictions are released under CC0 1.0. Compounds (identifiers, names, and desciptions) are from DrugBank, while diseases are from the Disease Ontology.