polypharmacy

A PyTorch and TorchDrug based deep learning library for drug pair scoring. (KDD 2022)

Readings for "A Unified View of Relational Deep Learning for Drug Pair Scoring." (IJCAI 2022)

TIP: Tri-graph Interaction Propagation model for Polypharmacy Side Effect Prediction (GRL@NeurIPS, 2019)

Application of Graph Neural Networks in accurate prediction of polypharmacy side effects.

Polypharmacy is a serious problem in this rapidly developing medical world where there are more types of medicines then the types of diseases. Due to this people are taking a lot of medicines together because of several health issues this is called polypharmacy and this result in various side effects. Our task to predict those side effects using Graph Neural Networks. As graphs are the closed thing to represent a drug molecule in computer science and do analysis on top of it. So using graph neural network we can find the correlation between drug-drug interations, protein-drug interactions and protein-protein interations.The Model takes all these interations into consideration and helps us to predict the side-effect probabilities due to drug-drug interaction. Here an edge between drug-drug corresponds to a side effect.

My first repository on GitHub

An easy-to-use Deep Learning tool for Polypharmacy Side Effect prediction.