AUTOPHARMAX

A multitask deep learning model that predicts cancer drug response and drug-pair synergy, with agentic tooling and a plain-language chatbot.

Multitask PyTorch model with dual MLP and graph neural network encoders predicting cancer drug response and drug-pair synergy, benchmarked against XGBoost and scikit-learn ensembles. Adds agentic tooling to query drug-efficacy data and a chatbot that turns model outputs into plain-language guidance (Streamlit).