Gene Identification Via Phenotype Sequencing: a Tool That Guides Sequencing Based Forward Genetics Studies

The cost of next generation sequencing (NGS) has been decreasing exponentially at a rate that beats the Moore's Law of computing hardware. Today, whole-genome sequencing has become a viable strategy for the identification of Mendelian genes from phenotype-exhibiting individuals. In such forward genetics studies, the choice of experimental parameters and analysis protocols has significant impact on the likelihood of successful gene identification. We present the Gene Identification via Phenotype Sequencing (GIPS) tool, which analyzes the impact of important experiment/analysis decisions, such as the number of individuals being sequenced, sequencing depth, sequence alignment/variance calling protocol, and variance selection strategy. For a specific experiment/analysis design, GIPS calculates three expected likelihoods: the chance of discovering the true phenotype-controlling gene, the chance of discovering unrelated genes, and the chance that the phenotype under scrutiny is controlled by more than one gene. GIPS calculates these likelihoods using both empirical data (e.g. genomic distribution of known Mendelian genes), and case-specific experimental data when they are available (e.g. the sequencing results of some phenotype-exhibiting individuals). These likelihoods may provide a theoretic guidance to effective planning and execution of NGS based forward genetics studies.