Integration and interpretation of multi-omics data are crucial for understanding complex diseases like cancer, COVID-19, and Alzheimer's disease.
Existing Graph Neural Networks (GNN) models are not well-suited for multi-omics data analysis.
The study introduces M3NetFlow, a novel GNN model that effectively integrates and interprets multi-omics data, achieving superior accuracy in drug combination prediction and aiding in personalized precision medicine. Additionally, a visualization tool called NetFlowVis is developed for analyzing drug targets and signaling pathways.