Fig. 6: Gene modules in AML patients. | Nature Communications

Fig. 6: Gene modules in AML patients.

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

Fig. 6

a Application of CGMega on AML. Multi-omics data of eight AML patients were obtained from a previous study. b In total, 396 candidate AML genes contained essential genes, transcription factors, and pan-cancer genes. Gray boxes show genes in two categories. c Gene ontology (GO) enrichments. We performed GO analysis on 597 known AML genes (first line), and on 993 genes (597 known AML genes and 396 candidate AML genes), respectively. GO analysis was conducted using DAVID and GO terms with p value lower than 1e-5 were shown. d Illustration of DLX4 gene module. Blue dots indicate known AML genes, while yellow dots indicate candidate AML genes. e Illustration of KLF4 gene modules in separate patients. SETD7 was located at the center in Patient 168 while it was just a participant in Patient 027 and Patient 270. In other patients, SETD7 did not appear in KLF4 gene modules. f We totally identified 142 neighbor-cancer gene pairs, these gene pairs were conserved in over four AML samples. Gene pairs in red box were detected in all eight AML samples and gene pairs in yellow box were detected in seven AML samples. Source data are provided as a Source Data file.

Back to article page