A research team led by Prof. ZHANG Guoqing from Shanghai Institute of Nutrition and Health of the Chinese Academy of Sciences has developed a knowledge-guided system that uses large language models to help scientists identify promising antigens for tuberculosis vaccines.

Tuberculosis (TB), caused by Mycobacterium tuberculosis (MTB), remains one of the world's deadliest infectious diseases. The Bacillus Calmette–Guérin (BCG) vaccine protects young children from severe TB, but provides limited protection against adult pulmonary TB. Developing new vaccines requires selecting protective antigens from nearly 4,000 MTB proteins. However, immunological evidence is scattered across a vast body of literature. Traditional computational tools mainly rely on sequence or structure features, while large language models (LLMs) alone may lack reliability and traceability.

To address this challenge, Prof. Zhang’s team, together with Prof. WANG Ying’s group from Shanghai Jiao Tong University School of Medicine, built MTB-ImmunogenKG, an LLM-assisted knowledge graph system. The researchers mined more than 77,000 publications indexed in PubMed and extracted 1.48 million sentence-level evidence records. The system integrates automated information extraction with knowledge-enhanced reasoning. It predicts antigen protective efficacy and traces supporting evidence. Importantly, it can detect contradictory findings reported in different studies and explain why experimental outcomes differ.

The system covers 3,154 MTB proteins, about 77% of annotated proteins in the genome. Compared with conventional sequence-based methods, it significantly improved prediction performance. It also outperformed a standalone LLM baseline. Moreover, the framework reveals antigen combinations and adjuvant associations that may influence immune protection. By organizing fragmented literature into structured knowledge, the system enables transparent and explainable decision-making in vaccine design.

The study was published in Biosafety and Health under the title "MTB-ImmunogenKG: An LLM-assisted knowledge graph for antigen selection in tuberculosis vaccine research" on Feb.4, 2026.

The study was supported by the National Key R&D Program of China, the Shanghai Science and Technology Innovation Action Plan, and the Guangzhou National Laboratory research program.

Paper link: https://doi.org/10.1016/j.bsheal.2026.02.001

Construction of MTB-ImmunogenKG. (Image provided by Prof. Zhang's team)

Scientific Contact:
Prof. ZHANG Guoqing
Shanghai Institute of Nutrition and Health,
Chinese Academy of Sciences
Email: gqzhang@sinh.ac.cn

Media Contact:
WANG Jin
Shanghai Institute of Nutrition and Health,
Chinese Academy of Sciences
Email: wangjin01@sinh.ac.cn