A molecular map of the living human brain from quantitative MRI

Profiling dynamic molecular processes in neurodegeneration in vivo remains a major clinical challenge. Limited brain tissue accessibility especially limits the development and effective deployment of emerging disease modifying therapies, highlighting the need for non-invasive methods for profiling molecular disease processes. We introduce a non-invasive imaging framework that integrates ultra-high-resolution 7T quantitative MRI (qMRI) with spatially-resolved transcriptomics (SRT) to infer cell-type and pathway-specific molecular features within the cortical grey matter. We demonstrate that our method predicts from qMRI data disease-associated changes to cell type proportions and molecular pathways within the grey matter cortex which are corroborated by orthogonal datasets, including bulk RNA sequencing, single nucleus RNA sequencing and high-performance liquid chromatography. This proof-of-concept establishes the integration of qMRI and SRT as a scalable, non-invasive platform for molecular profiling in neurodegeneration, with significant potential for precision therapeutic monitoring, drug development and clinical trials.