Fig. 3From: PASNet: pathway-associated sparse deep neural network for prognosis prediction from high-throughput dataGraphical representation among the output layer, hidden layer, and pathway layer in PASNet. (a) The weights between the hidden layer and the output layer. Hidden nodes are sorted in a descending order. (b) The node values in the hidden layer. The horizontal dotted lines indicates LTS/non-LTS samples. The vertical dotted lines indicates LTS/non-LTS samples are significantly distinguished by top 16 pathways. (c) The absolute weights between the pathway layer and the hidden layerBack to article page