Skip to main content

Table 6 Average c-index values of the proposed model and random forest model using different numbers of randomly selected signaling pathways. The mean c-index was obtained by randomly selecting the training (80% of the dataset) and test (20% of dataset) data for 50 times

From: Investigating the relevance of major signaling pathways in cancer survival using a biologically meaningful deep learning model

# of pathway Proposed model Random forest model
Mean c_index on training data Mean c_index on test data Mean c_index on training data Mean c_index on test data
GBM
10 0.6428 0.6115 0.6832 0.5182
20 0.6442 0.6129 0.672 0.5159
30 0.6247 0.5975 0.6617 0.5314
40 0.6300 0.6196 0.6602 0.5418
SKCM
10 0.6948 0.6626 0.7684 0.4943
20 0.7226 0.6702 0.7701 0.4434
30 0.7197 0.6614 0.772 0.4336
40 0.7629 0.6772 0.7739 0.4265
BRCA
10 0.4397 0.3713 0.7859 0.5294
20 0.6051 0.4577 0.8029 0.5225
30 0.6661 0.5622 0.81 0.5282
40 0.6497 0.5506 0.8093 0.5139
LUAD
10 0.7349 0.7050 0.8158 0.5195
20 0.7226 0.7131 0.8162 0.5349
30 0.7347 0.7171 0.8101 0.5551
40 0.7119 0.7226 0.8091 0.5669