Multikingdom microbiome-based machine learning enables multiple sclerosis diagnosis
Gaochen Zhu, Guan Yang · 2026
Read Multikingdom microbiome-based machine learning enables multiple sclerosis diagnosis by Gaochen Zhu, Guan Yang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13519030.txt.
Description
Gaochen Zhu and Guan Yang built a stool-based test for multiple sclerosis that reads the whole gut microbe community, not just bacteria. Published in 2026, the study analyzed gene-sequence data from stool samples, with 1,152 people in the discovery group and 1,306 in total across three regions. Machine learning models then learned to separate patients from healthy controls using microbes from several kingdoms plus their genes and metabolic pathways.
The models flagged 90 bacterial species, a handful of fungal and viral species, 119 gene families, and 17 metabolic pathways tied to the disease. In the discovery data the models scored near the top of the accuracy scale for both men and women, though scores fell on outside test sets. A shorter 30-marker version held up better externally. Low levels of Faecalibacterium microbes and a methionine-making pathway tracked with the disease. The work points toward a noninvasive stool marker, still needing wider validation.
Catalog record
| CREATOR | Gaochen Zhu, Guan Yang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |