Summary

Drug-drug side-effects are rarely tested directly in the pharmaceutical development process. Our system can generate in silico indications of drug-drug interactions using graph machine learning.

Overview

Identifies Drug interactions and potential side effects using graph machine learning during the pharmaceutical development process.
TigerLily
Drug Interaction Prediction with Tigerlily

Tigerlily is a TigerGraph based system designed to solve the drug interaction prediction task. In this machine learning task, we want to predict whether two drugs have an adverse interaction. Our framework allows us to solve this highly relevant real-world problem using graph mining techniques in these steps:

  1. Using PyTigergraph we create a heterogeneous biological graph of drugs and proteins.
  2. We calculate the personalized PageRank scores of drug nodes in the TigerGraph Cloud.
  3. We embed the nodes using sparse non-negative matrix factorization of the personalized PageRank matrix.
  4. Using the node embeddings we train a gradient boosting based drug interaction predictor