Skip to main content

Table 4 The size of encoder layers and the 10-fold cross-validation accuracy of each neural network classifier

From: GSAE: an autoencoder with embedded gene-set nodes for genomics functional characterization

NN classifiera Input
Type
Encoder Layer
1d
Encoder Layer
2
Encoder Layer
3
Encoder Layer
4
Accuracy of 10-fold cross validation
Superset Genesb 2334 200    88.79%
Gene set Genes 2334     87.69%
2-layer fc Genes 2334 200    47.86%
2-layer fc Genes 2000 500    37.98%
4-layer fc Genes 2000 200 100 50 46.06%
2-layer fc PCc 400 100    87.57%
4-layer fc PC 200 200 100 25 87.57%
  1. a2-, 4-layer fc: 2- or 4- layer fully connected AE
  2. bGenes input is the 15,183 genes of TCGA BRCA RNA-seq data
  3. cPC input is the top 500 principal components of PCA analysis
  4. dThe encoder layer 1 of superset and gene set classifier is the gene set layer (not a fully connected layer)