Longevity & AgingAI Maps ME/CFS Biology Linking Gut Microbes, Immunity and Metabolites to Symptoms
Researchers developed BioMapAI, a deep neural network trained on a 4-year longitudinal multi-omics dataset from 249 participants (153 ME/CFS patients, 96 healthy controls). The model integrates gut metagenomics, plasma metabolomics, immune cell profiling, blood labs, and clinical symptom scores to classify disease and identify symptom-specific biomarkers. Using explainable AI (SHAP), the team constructed a connectivity map revealing that depletion of microbial short-chain fatty acids and branched-chain amino acids in ME/CFS drives abnormal activation of MAIT and γδT cells producing IFN-γ and GzA, while reduced microbial tryptophan and benzoate connections to plasma lipids correlate with fatigue, emotional dysregulation, and sleep disturbances. BioMapAI achieved 91% AUC in distinguishing ME/CFS from healthy controls.
