bioRxivJClub
An atlas of human histological diversity
bioRxiv · · Preprint · Open access
Mantion, Zhang + 4 more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
An unsupervised histology atlas built from GTEx links 11,125,747 associations between 2,560 morphemes and 9 molecular/patho layers across 40 organs from 946 non-diseased individuals.
- Why it matters: Histopathology descriptions are constrained by verbal, biased, and quantitatively inaccurate assessments, limiting unbiased interpretation of tissue structure.
- What they did: The study adapts RNA-seq-style unsupervised computational analysis to histology, using GTEx to map histological morphemes to multi-omic and patho-clinical data layers.
- The result: The atlas shows age, sex, genetics, and sub-clinical pathologies shape tissue structures, including sex-linked adipocyte differences around female eccrine sweat glands and calcification-associated aorta morphemes linked to smoking or genetic polymorphisms.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access