Fig. 4: Mean phenotype similarity evaluations in the reference and other datasets.
From: A systematic analysis of mitochondrial aminoacyl tRNA synthetase variants in a rare disease cohort
A The mean phenotype similarity scores per person, by gene, across the mt-aaRS reference dataset. B Receiver operating characteristic (ROC) plot using the training and test datasets for modelling the mean phenotype similarity score to detect individuals with mt-aaRS-related diseases. Area under the curve (AUC) values and the corresponding confidence intervals (CI) are displayed on the graph for the balanced and unbalanced datasets evaluated through the generalised linear (GLM) and random forest (RF) models.
