Single-nucleotide polymorphisms (SNPs) are among the most prevalent forms of genetic variation and are frequently associated with human diseases. In this study, we performed a comprehensive in silico analysis of non-synonymous SNPs (nsSNPs) in the GUSB gene, which encodes the lysosomal enzyme $β$-glucuronidase, a key regulator of glycosaminoglycan degradation. A total of 449 reported mutations were systematically evaluated using sequence- and structure-based predictive algorithms. Among these, 15 variants were identified as deleterious and structurally destabilizing. Subsequent pathogenicity assessment using SNPs&GO, MutPred, and PhD-SNP further narrowed these to eight high-confidence pathogenic mutations. Solubility and aggregation propensity analysis using the SODA tool revealed that 37.5% of these pathogenic variants showed increased aggregation or reduced solubility, suggesting a potential mechanism contributing to disease progression. Detailed structural investigation indicated that the observed destabilization likely arises from alterations in interatomic non-covalent interactions, ultimately compromising protein stability and function. Overall, this study provides a systematic and integrative characterization of pathogenic nsSNPs in the GUSB gene. The findings enhance our understanding of the structural and functional consequences of these variants and offer a foundation for future experimental validation and the development of targeted therapeutic strategies.