Family-based integrative genomics identifies neural–metabolic–skeletal pathways underlying scoliosis susceptibility

Abstract

Scoliosis is a complex spinal deformity with strong familial aggregation, yet its genetic basis remains incompletely defined. Here, we performed whole-exome sequencing (WES) in 34 scoliosis families, identifying candidate single nucleotide polymorphisms (SNPs) potentially contributing to disease susceptibility. To connect genomic variation with transcriptional regulation, we integrated these data with cis-expression quantitative trait loci (cis-eQTL) summary statistics from eQTLGen and scoliosis GWAS data from FinnGen R12 using summary-based Mendelian randomization (SMR). After harmonization and filtering (P_SMR < 0.05, P_HEIDI > 0.05), we identified 31 shared genes overlapping between family-based SNPs and eQTL-associated loci. Among them, LINC00299 and NPIPB11 exhibited strong colocalization signals, with LINC00299 expression inversely correlated with scoliosis risk. Functional annotation further highlighted NCAM1, AGPAT3, SMIM12, and CTSH as core pathogenic nodes linking neural development, lipid metabolism, mitochondrial energy regulation, and bone remodeling. Together, these findings reveal a multi-systemic genetic architecture of scoliosis and nominate metabolic-neuro-skeletal pathways as potential targets for precision diagnosis and therapy.