He, Y., Zheng, Y.
A framework that leverages biomedical ontologies and knowledge graphs to solve the vaccine adjuvant recommendation problem using a graph neural network.
Selecting an effective adjuvant remains a bottleneck in vaccine development, and most existing computational efforts have targeted antigen discovery rather than adjuvant prioritization.
The study frames disease-adjuvant matching as a top-k recommendation task on a heterogeneous knowledge graph grounded in biomedical ontologies, integrating curated facts, mechanistic pathways, and textual evidence. It introduces VaxjoGNN, a graph neural network trained with a listwise ranking objective.
On a public benchmark, VaxjoGNN achieves NDCG@10 of 0.59 on seen diseases and 0.27 on previously unseen diseases (a 5.4x improvement over a random baseline). The framework provides an ontology-anchored approach to adjuvant prioritization that complements existing antigen-focused tools.