Abstract
Skeletal dysplasias represent a heterogeneous group of genetic disorders characterized by abnormal bone development. Despite known primary genetic causes, phenotypic severity varies widely among patients with identical mutations, suggesting the influence of modifier genes. This study integrates multi-omics data—genomics, transcriptomics, and proteomics—from a cohort of 150 individuals with skeletal dysplasias to identify genetic modifiers of disease severity. Using a similarity regression fusion model [1] and random forest integration [8], we analyzed whole-exome sequencing, RNA-seq from bone biopsies, and bone proteomics profiles. Patients were stratified into mild (n=72) and severe (n=78) phenotypes based on clinical scoring. The integrated analysis identified 12 candidate modifier genes significantly associated with severity after multiple testing correction (FDR
Keywords
Multi-Omics Integration, Skeletal Dysplasia, Modifier Genes, Phenotypic Severity, Systems Biology, Cohort Study, Machine Learning, Bone Development