Familial hypercholesterolemia (FH) is a metabolic disorder that leads primarily to premature cardiovascular diseases, the main cause of mortality in Saudi Arabia (SA). FH is underreported and underdiagnosed in SA with statistical evidence of high expected prevalence in such a consanguineous community. Lacking knowledge of which and how these alterations are actually impacting lipid metabolism is one of the main reasons why FH is insufficiently diagnosed in the region. The aim of this study was to develop a fast prediction approach using an integrated bioinformatics method for future screening of the potential causative variants from national registries. A total of 21 variants were detected with majority rate in LDLR (81%). Variants were classified based on the type of mutation. Missense variants resulting in amino acid changes, c.1429G>A (p.D477N), c.1474G>A (p.D492N), c.1731G>T (p.W577C), and c.1783C>T (p.R595W) in LDLR gene, in addition to c.9835A>G (p.S3279G) in APOB, were shown to be deleterious by concordant analysis. Furthermore, functional interaction deformities showed a significant loss and gain of energies in the mutated proteins. These findings will help in distinguishing the most harmful mutations needed to be screened for clinically diagnosed FH patients in SA. Such computational research is necessary to avoid time consumption and the usage of expensive biological experiments. This can be a fast track to facilitate the future filtering and screening of causative mutations from national registries.
Keywords: bioinformatic; computational analysis; familial hypercholesterolemia; in silico; mutation.