Table S4 from Deep-Learning Model for Tumor-Type Prediction Using Targeted Clinical Genomic Sequencing Data

Stepwise accuracy of various models generated during model development, including direct comparison of GDD-RF to GDD-ENS. Acc. = Accuracy, Conf. = Confidence, Macro Prec: Class-Averaged Precision, % Excluded: proportion of high-content, solid tumor training data that is not represented by the model training types. RF: Random Forest. GDD-RF Feature Set, training set and testing set = samples and features used in original GDD-RF model. GDD-ENS feature and training set = updated samples and features used in final GDD-ENS model. GDD-ENS-22 test set: GDD-ENS test set but limited to the 22 original types before expansion. GDD-RF-EXT test set = extended independent test set used in GDD-RF training. GDD-ENS-38 = Final GDD-ENS test set.