Unveiling genetic architecture of white matter microstructure through unsupervised deep representation learning of fractional anisotropy images

Fractional anisotropy (FA) derived from diffusion MRI is a widely used marker of white matter (WM) integrity. However, conventional FA-based genetic studies focus on phenotypes representing tract- or atlas-defined averages, which may oversimplify spatial patterns of WM integrity and thus limit the genetic discovery. Here, we proposed a deep learning-based framework, termed unsupervised deep representation of WM (UDR-WM), it adopted the voxel-wise FA maps as the input, and to extract brain-wide FA features--referred to as UDIP-FA--that capture distributed microstructural variation without prior anatomical assumptions. UDIP-FAs exhibit enhanced sensitivity to aging and substantially higher SNP-based heritability compared to traditional FA phenotypes (P < 2.20x10-16, Mann-Whitney U test, mean = 50.81%). Through multivariate GWAS, we identified 939 significant lead SNPs in 586 loci, mapped to 3480 genes, dubbed UDIP-FA related genes (UFAGs). UFAGs are overexpressed in glial cells, particularly in astrocytes and oligodendrocytes (P < 8.03x 10-8, Wald Test), and show strong overlap with risk gene sets for schizophrenia and Parkinsons disease (P < 1.10 x 10-4, Fisher exact test). UDIP-FAs are genetically correlated with multiple brain disorders and cognitive traits, including fluid intelligence and reaction time, and are associated with polygenic risk for bone mineral density. Network analyses reveal that UFAGs form disease-enriched modules across protein-protein interaction and co-expression networks, implicating core pathways in myelination and axonal structure. Notably, several UFAGs, including ACHE and ALDH2, are targets of existing neuropsychiatric drugs. Together, our findings establish UDIP-FA as a biologically and clinically informative brain phenotype, enabling high-resolution dissection of WM genetic architecture and its genetic links to complex brain traits.