G.P.286: Validation of in silico variation effect prediction tools in missense mutations of dysferlinopathy

Abstract

In silico prediction tools for genetic variation have now become the commodity in genetic analysis. Each tool utilizes their own algorithm to give us the results. Dysferlin (DYSF) is a transmembrane protein, crucial for sarcolemmal repair, and its recessive genetic defects bring about progressive muscular weakness either as Miyoshi distal myopathy or limb girdle muscular dystrophy 2B. The Leiden database lists one of the most corroborating data set on the pathogenicity of each genotype in the field of muscular dystrophy. In this study, we compared the in silico analysis results and the published pathogenicity of the dysferlin mutations listed on Leiden database. We suggest a new cut-off value that can be optimally used to test a new mutation with in silico tools. We used PROVEAN and SIFT online software to predict a possibly damaging mutation. This is a prediction tool used mainly to test the impact on the biological function of a protein by a missense mutation or amino acid substitution. We used the Leiden open variation database on dysferlin mutations and utilized the complete sequence variants data provided by the website. PROVEAN tool showed a remarkably better result to test dysferlin nonsense mutations than SIFT tool. On using PROVEAN tool for dysferlin missense mutation, we propose a new cut-off value of −5.365. Individualized cut-off values for other genes will expand the usefulness of in silico prediction tools.