Supplementary Figure S1.

Performance of model neural network interpretability A. Spearman’s correlation on gene contribution scores and (B) overlap in the top 100 contributing genes in lung (upper panel), ovary (middle panel), and pancreas (lower panel) among replicates from the same pre-trained model, different pre-trained models with the same gene order, and different pre-trained models with different gene orders based on CNN. C. Spearman’s correlation on gene contribution scores and D. overlap in the top 100 contributing genes in lung (upper panel), ovary (middle panel), and pancreas (lower panel) among replicates from the same pre-trained model and different pre-trained models based on MLP. CNN, convolutional neural network; MLP, multilayer perceptron.