Predicting the Pathogenic Effects of Non-Synonymous Single Nucleotide Variants in Human Diseases

Single nucleotide variants (SNVs) are a common type of genetic variants, defined as single base substitutions occuring at a specific position in the genome, and can be categorized as synonymous and non-synonymous. Non-synonymous SNVs (nsSNVs), also known as missense variants, are mutations in the coding region of the genome leading to an alteration in the amino acid sequence. Several computational tools and algorithms have been developed to predict the pathogenic effects of such variants. These methods rely on diverse methods and feature sources including sequence conservation, sequence features, physico-chemical properties of amino acids and functional annotations to assess the potential functional impacts of nsSNVs. A few computational methods also incorporate the features derived from 3D structural information of proteins to compute deleterious scores and predict the pathogenicity of missense variants. These prediction tools are mainly classified as sequence-based, structure-based, and sequence and structure-based methods that integrates information from both sequence and structure. Ensemble methods combine the results of multiple tools to improve the performance of pathogenicity prediction. There has been major developments in computational tools for predicting the pathogenicity of nsSNVs in order to support clinical variant interpretation, but at the same time these tools are sometimes limited in generating accurate and meaningful predictions of potential causal variants. Despite advances in pathogenicity prediction methods, the classification of missense variants as pathogenic or benign remains a major challenge. The challenge lies in the complexity of biological systems and the inherent inter-individual genetic variability which contributes to the difficulty in reliable predictions of the pathogenicity of nsSNVs. This chapter provides a brief overview of the tools and methods available for predicting the pathogenicity of nsSNVs. We also provide a few examples from studies illustrating the use of these tools for predicting missense variants in human diseases.