IBDome: An Integrated Molecular, Histopathological, and Clinical Atlas of Inflammatory Bowel Diseases.

In a multicohort study of 1002 clinically annotated patients, site-specific inflammatory signatures distinguish Crohn’s disease and ulcerative colitis in normal versus inflamed gut tissues.

• Why it matters: Multiomic and multimodal datasets with detailed clinical annotations are needed to refine diagnostics and support personalized therapeutic strategies in inflammatory bowel diseases.
• What they did: The study integrated whole-exome and RNA sequencing of normal and inflamed gut tissues, serum proteomics, and histopathology from H&E images, then built a serum inflammatory protein severity signature and foundation model deep learning.
• The result: Serum proteomics produced an inflammatory protein severity signature linked to intestinal molecular inflammation, while deep learning accurately predicted histologic disease activity scores and classified Crohn’s disease versus ulcerative colitis from H&E images.

AI summary of the abstract · may contain errors · not medical advice · read the paper.