Skip to main content

Advertisement

Table 4 Area under ROC curve (AUROC) and Area under PR curve (AUPR) for gene Interaction Prediction

From: GNE: a deep learning framework for gene network inference by aggregating biological information

Methods Yeast E. coli
  AUROC AUPR AUROC AUPR
Correlation 0.582 0.579 0.537 0.557
Isomap 0.507 0.588 0.559 0.672
LINE 0.726 0.686 0.897 0.851
node2vec 0.739 0.708 0.912 0.862
Isomap+ 0.653 0.652 0.644 0.649
LINE+ 0.745 0.713 0.899 0.856
node2vec+ 0.751 0.716 0.871 0.826
GNE (Topology) 0.787 0.784 0.930 0.931
GNE (our model) 0.825* 0.821* 0.940* 0.939*
  1. + indicates the concatenation of expression data with learned embeddings to create final representation. * denotes that GNE significantly outperforms node2vec at 0.01 level paired t-test. Note that method that achieves the best performance is bold faced