Huaiwu Zhang - A Causal Inference Framework for Identifying Essential Genes to Enhance Drug Syn Pred

Finnish Society for Bioinformatics Webinar Series 2026 Speaker: Huaiwu Zhang, PhD Candidate in Bioinformatics Affiliation: University of Helsinki, Finland Topic: A Causal Inference Framework for Identifying Essential Genes to Enhance Drug Synergy Prediction Date: 10th September 2026, Thursday 14:00 (Helsinki Time/EET) Abstract: Identifying synergistic drug combinations may enable more effective treatments. However, most deep learning methods do not explicitly model the causal effects of genes on drug responses. We introduce CADS (Causal Adjustment for Drug Synergy), a framework that integrates multi-omics data with learnable gene selection and causal backdoor adjustment. CADS supports both synergy prediction and causal gene discovery. Experiments show that it outperforms state-of-the-art models, while case studies confirm that its causal scores recover clinically validated cancer-related genes. These findings demonstrate that causal modelling improves the accuracy and interpretability of drug synergy prediction.