Can Chen: Hyperlink prediction in biological networks

Abstract: Various biological networks, such as metabolic reactions, drug interactions, and gene pathways, exhibit intricate higher-order interactions. However, these networks are often incomplete and poorly characterized due to limitations in current knowledge and technology. Traditional graph representations, which primarily capture pairwise relationships, fall short of capturing the full complexity of these interactions. Hyperlink prediction provides a powerful framework for uncovering missing higher-order interactions by modeling these networks as hypergraphs, a generalization of graphs where hyperlinks can connect multiple nodes simultaneously. In this talk, I will discuss recent advancements in hyperlink prediction and its applications to genome-scale metabolic networks, drug interactions, and gene pathways in bacteria.