Skip to content

Clinical Utility and Genes of Interest

Jim Havrilla edited this page Mar 16, 2018 · 1 revision

By default, the top ten genes ranked by number of submitted variants in ClinVar are provided for the user courtesy of Flygare et al. (2018). For those who are unfamiliar, the clinical utility score done by gene for each metric, and is calculated by first discovering the metric value with the peak J-score (Youden statistic) for a metric, i.e. J = specificity + sensitivity - 1. Then everything above that value is called a positive, everything below it is called a negative. The accuracy, or (TP+TN)/(TP+TN+FP+FN), is calculated from that for all variants that can be scored for a particular gene and that is then multiplied by the number of variants that were able to be scored by that metric of all total variants for the gene of interest. The score ranges from 0 to 1 with 1 being the ideal score.

Clone this wiki locally