Naafey Aamer

This paper introduces GNN-GBA, a graph neural network that identifies mechanistic pathways linking gut microbes to brain disorders using a large biomedical knowledge graph.

COMIC improves drug repurposing prediction with explainable pathways, achieving 9.55% better performance than existing methods on a large knowledge graph.

BIND predicts biological interactions using knowledge graph embeddings and machine learning, achieving high F1 scores and identifying novel drug-phenotype interactions supported by literature.

This paper introduces a framework combining knowledge graphs and LLM agents to automate biomedical discovery by identifying relevant pathways, designing AI predictors, and proposing wet-lab experiments.