Hemma Murali, B.A. [1,2]; Kim Sharp, PhD [2], Eric Liao, MD, PhD [3]; Kai Wang, PhD [1] 1. Department of Pathology and Lab Medicine, University of Pennsylvania 2. Department of Biochemistry and Molecular Biophysics, University of Pennsylvania 3. Division of Plastic, Reconstructive and Oral Surgery, Children's Hospital of Philadelphia
Poster # 41
Next-generation sequencing revolutionized genetic testing, revealing numerous rare disease-associated variants. However, many variants are classified as unactionable Variants of Uncertain Significance (VUS) since the volume and complexity of these data pose challenges to determining the clinical significance. Existing in silico variant pathogenicity prediction tools tend to overlook the impacts of variants on 3D protein structure since experimental structures for many proteins are unavailable. The emergence of AlphaFold2 has transformed the field of protein structure determination, so we outlined a strategy that leverages structure prediction programs and genetic variant databases to enhance VUS reclassification. We used the gene IRF6 as a case study since previous studies have experimentally validated mutations through phenotype rescue experiments in irf6-/- zebrafish. We compared results from over 30 pathogenicity prediction tools on 37 IRF6 variants. IRF6 lacks an experimentally derived structure, so we used predicted structures to explore correlations between mutational clustering and pathogenicity. We found that among these IRF6 variants, 19 of 37 were unanimously predicted as deleterious by computational tools. Comparing in silico predictions with experimental findings, 12 variants predicted as pathogenic were experimentally determined as benign. Mapping variants to the protein revealed deleterious mutation clusters around the protein binding domain. Interestingly, N-terminal variants showed contrasting tool-to-experiment results. In general, these tools favor labeling variants as pathogenic, highlighting the need for biological considerations in current predictive tools. In conclusion, incorporating structural features from proteins and analyzing mutation neighborhoods may enhance classification and provide meaningful insights into pathogenicity predictions.
Many training centers in Coimbatore boast state-of-the-art infrastructure, including modern classrooms, high-speed internet, and advanced computing facilities. This conducive learning environment python training centres in coimbatore enhances the educational experience, allowing students to focus on their studies without any technical hindrances. The availability of the latest software and tools further aids in providing up-to-date knowledge and skillsA significant advantage of enrolling in a Python training center in Coimbatore is the robust placement assistance and career support provided.