Quantitative imaging of native tissue architecture combined with deep multi-omics reveals the spatial biology of kidney aging

NORIGAMI predicts chronological age in held-out male rat kidneys from label-free NoRI images with mean absolute error 0.78 months (R² = 0.97), outperforming tubule-only models.

• Why it matters: Structural and compositional changes during renal aging are spatially heterogeneous, but how these measurable tissue phenotypes connect to underlying molecular programs is unclear.
• What they did: In 32 male rats aged 9, 18, 21, and 25 months, the pipeline integrates NoRI with deep learning, quantitative morphology, deep proteomics (11,485 proteins), and metabolomics (269 metabolites).
• The result: It identifies 41 age-associated morphological/compositional features (FDR < 0.05) and a protein clock (mean absolute error 1.6 months), linking NAD+ associated molecular shifts to tubular geometry and protein–lipid organization.

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