Predicting Drug-Drug Interactions with Hypergraphs — on Two CPU Cores

Automated excerpt

Every distinct substring becomes a node, and every drug becomes a hyperedge connecting the substrings it contains. We did something else: each of the 86 interaction types becomes a hyperedge, connecting every drug that participates in that type. This hypergraph is tiny — 1,709 drug nodes, 86 type hyperedges, 13,486 connections. Hyperedge sizes range wildly: type 42 touches 5 drugs, type 49 touches 1,308. Attention then flows hyperedge → node → hyperedge, so the type hyperedges — which are the prediction target — end up holding the final aggregated representation.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.