Multi-omics characterization of white matter free water reveals genetic and vascular determinants of neurological disorders

White matter free water (FW-WM), quantified using diffusion MRI, reflects extracellular water content and microstructural alterations within water matter, but its genetic architecture and clinical relevance remain largely uncharted. We conducted a multi-omics study integrating 42,039 individuals from the UK Biobank and Alzheimer’s Disease Neuroimaging Initiative cohorts to characterize tract-specific FW-WM across genetic, phenotypic, and proteomic dimensions. Genome-wide association analyses identified 27 independent loci and prioritized 88 genes, which were enriched in the genomic functional elements of vascular endothelial cells, pericytes, and astrocytes. Notably, 20 loci (74%) were novel compared to conventional white matter metrics. Both genetic predisposition to higher FW-WM and imaging-derived FW-WM were associated with a spectrum of neurological disorders. Specifically, imaging-derived FW-WM showed strong associations with stroke (OR = 3.14) and dementia (OR = 4.15), while higher baseline FW-WM predicted progression from MCI to AD (HR = 2.58). Bidirectional Mendelian randomization analyses further supported genetically predicted associations between hypertension and FW-WM, which were together associated with increased risks of stroke and cognitive impairment. Integrative proteomic analyses further nominated PEAR1 and ANGL3 as candidate proteins potentially linking hypertension-related vascular processes to FW-WM variation. Together, these findings provide a multi-omics characterization of FW-WM that connects genetic susceptibility, vascular factors, and neurological outcomes, supporting FW-WM as a tract-specific imaging phenotype with potential utility for neurological risk stratification.