Yagin, Fatma Hilal, Korkmaz, Yavuz, Colak, Cemil et al. · International journal of molecular sciences · 2026 · DOI
Researchers analysed blood metabolite patterns (small molecules in blood) from 106 people with ME/CFS and 91 healthy controls using machine learning to see whether these patterns could distinguish ME/CFS from health. A computer model called Explainable Boosting Machine achieved 91% accuracy in this one study group. Importantly, the findings remain preliminary—they describe associations observed in this cross-sectional sample and require external validation in new populations before clinical use.
ME/CFS lacks objective diagnostic biomarkers. This study demonstrates that metabolomic signatures—specifically patterns of co-variation between metabolites—are associated with ME/CFS classification in a single cohort, potentially informing future diagnostic strategies. The use of explainable machine learning provides transparency about which metabolite relationships contributed most to classification, bridging the gap between high-dimensional data and clinical interpretability.
This study does not establish causation or confirm that metabolomic dysregulation drives ME/CFS pathogenesis. It does not prove the identified metabolite interactions are disease mechanisms rather than biomarkers of other unmeasured processes. Cross-sectional design means temporal direction is unknown. The findings have not been externally validated in independent populations and remain unproven for clinical diagnostic use. Peer-review status is unknown.
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 →
The first block is for the primary paper and is the citation you should use in research work. The atlas-snapshot line only applies if you are specifically referring to this atlas’s reading of the paper on the date shown.
Primary citation
Yagin, Fatma Hilal, Korkmaz, Yavuz, Colak, Cemil, Alzakari, Sarah A, Alkhalifa, Amal K, Al-Hashem, Fahaid, et al. (2026). Metabolomic Classification of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome via Explainable Ensemble Learning and Pareto-Guided Feature Selection.. International journal of molecular sciences. https://doi.org/10.3390/ijms27135920
BibTeX
@article{mecfsatlas-yagin-2026-metabolomic-classification,
author = {Yagin, Fatma Hilal and Korkmaz, Yavuz and Colak, Cemil and Alzakari, Sarah A and Alkhalifa, Amal K and Al-Hashem, Fahaid and Aghaei, Mohammadreza},
title = {Metabolomic Classification of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome via Explainable Ensemble Learning and Pareto-Guided Feature Selection.},
journal = {International journal of molecular sciences},
year = {2026},
doi = {10.3390/ijms27135920},
note = {PubMed: 42450188},
url = {https://www.mecfsatlas.com/evidence/yagin-2026-metabolomic-classification},
}Atlas snapshot reference
ME/CFS Atlas. Generator v1 / Scanner v1.4 / policy v0.1. Accessed 2026-07-16. https://www.mecfsatlas.com/evidence/yagin-2026-metabolomic-classification
Contribute
Private, reviewed by a human. Not a public comment thread.