Genomic structural equation modeling reveals a common genetic dimension contributing to variation in growth‑related traits across multiple developmental stages

Adult height serves as a comprehensive biological proxy for early-life conditions and long-term well-being. Despite the well-established genetic basis of height, current research predominantly employs single-trait or timepoint-specific analyses, thereby limiting the systemic elucidation of the shared genetic architecture spanning from developmental initiation to structural maturation. We applied Genomic Structural Equation Modeling (Genomic SEM) to integrate large-scale GWAS summary statistics for five phenotypes: birth weight, IGF-1 levels, bone mineral density, childhood height, and adult height. Post-GWAS analyses included fine-mapping, TWAS with FOCUS, pathway enrichment, cell-type specificity, and spatial transcriptomic mapping. A single latent growth trait demonstrated excellent model fit (CFI = 0.996, SRMR = 0.028). Multivariate GWAS identified 2,120 independent lead variants, with fine-mapping prioritizing high-confidence causal variants at loci including ACAN, IHH, and ADAMTS17. Pathway enrichment converged on endochondral ossification and GH/IGF-1 signaling. Cell-type analysis revealed significant enrichment in mesenchymal lineages, while spatial mapping localized signals predominantly to the cartilage primordium. This study provides a unified genetic framework connecting early-life initiation, molecular mediation, and structural development, offering novel insights into the systemic developmental principles governing human height.