Fig. 2: CGMega performance in cancer gene prediction task. | Nature Communications

Fig. 2: CGMega performance in cancer gene prediction task.

From: CGMega: explainable graph neural network framework with attention mechanisms for cancer gene module dissection

Fig. 2

a AUPRC on breast cancer cell line MCF7. b Methods comparison on MCF7 cell line. N2V represents node2vector. MLP and SVM were tested with and without (w/o) PPIs. c AUPRCs of non-pretrained CGMega and pretrained CGMega on datasets with different numbers of labeled genes. d Ablation experiments. AUPRCs of CGMega with random or without omics features and Hi-C features. e AUPRC of CGMega on Hi-C data under different resolutions and down-sampling rates. Source data are provided as a Source Data file.

Back to article page