Spatiotemporal AI modeling of gene regulation and variant effects in the developing human brain

Abstract

Understanding how genetic variation shapes gene regulation in the developing human brain remains a central challenge in neuroscience, particularly in transient cell types and developmental stages where data are limited. We present Seal, an interpretable AI transfer learning framework for genome-based modeling of gene expression and variant effects with spatiotemporal resolution across 26 brain regions, 30 cell types, and seven developmental stages. Seal predictions enable characterization of both germline and somatic brain variants, and provide mechanistic interpretation by linking sequence variants to transcriptional regulators in spatiotemporal context. Applying Seal to genome-wide association studies, we identify cell types and developmental windows relevant to neuropsychiatric disease risk, revealing shared and distinct regulatory architectures across six neuropsychiatric conditions that align with clinical trajectories. In the Simons Simplex Collection autism cohort whole genome data, Seal uncovers a significant burden of de novo regulatory variants in transient fetal excitatory neurons, revealing a cell-type and developmentally specific regulatory burden. By connecting genetic variation to spatiotemporal regulatory programs, Seal offers a general framework for uncovering mechanisms of human brain development and disease.