Fig. 4From: Off-targetP ML: an open source machine learning framework for off-target panel safety assessment of small moleculesLollipop chart representing the balanced accuracy of each target model for each method. Each subplot represents one of the methods used to build the models (Neural networks, Auto-Sklearn, AutoGluon, H2O and Random Forest). The gene names/abbreviations of each target model are represented on the x-axis and the corresponding balanced accuracy values on the y axis. Lines are color coded according to the targets’ hit percent category and the numerical values of the hit percent are indicated inside the circlesBack to article page