Xiong, Ruoyun, Aiken, Elizabeth, Caldwell, Ryan et al. · Nature medicine · 2025 · DOI
Researchers used artificial intelligence to analyze blood, gut bacteria, immune cells, and symptoms from 249 ME/CFS patients tracked over 4 years. The AI model discovered that ME/CFS involves abnormal patterns in gut bacteria products, blood fats, and immune cells that attack infections—particularly special T cells that become overactive. These findings suggest ME/CFS isn't caused by a single problem but by multiple connected systems going wrong together.
This is the first systems-level computational analysis linking microbiome, metabolic, and immune dysfunction in ME/CFS, moving beyond single-biomarker studies. It provides a framework for understanding why ME/CFS presents with heterogeneous symptoms and may eventually enable personalized diagnostic and treatment strategies. The explainable AI approach makes complex biological networks interpretable for both researchers and clinicians.
This study does not prove that altered bacteria, metabolites, or immune cells cause ME/CFS—only that they are associated with the disease. It cannot establish the direction of causality (whether immune dysregulation drives microbiome changes, or vice versa). The findings require independent replication and functional validation before they can inform clinical practice.
About the PEM badge: “PEM required” means post-exertional malaise was an explicit required diagnostic criterion for participant inclusion in this study — not that PEM was studied, observed, or discussed. Studies using criteria that do not require PEM (e.g. Fukuda, Oxford) are tagged “PEM not required”. How the atlas works →
Contribute
Private, reviewed by a human. Not a public comment thread.