Prediction and interpretation of cancer survival using graph convolution neural networks

• Published in: Methods (San Diego, Calif.) Vol. 192; pp. 120 - 130
• Main Authors: Ramirez, Ricardo, Chiu, Yu-Chiao, Zhang, SongYao, Ramirez, Joshua, Chen, Yidong, Huang, Yufei, Jin, Yu-Fang
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.08.2021
• Subjects: Algorithms; brain; breast neoplasms; Breast Neoplasms - genetics; breasts; colorectal neoplasms; data collection; genes; Graph convolutional neural network; Humans; Male; Neural Networks, Computer; prediction; risk; Survival analysis; Survival Rate; testes; The cancer genome atlas (TCGA); PRAD; ANN; RS; CNN; TGCT; UCS; UCEC; UVM; C-index; LUAD; PCPG; COAD; MESO; Cox-PH; GBM; The cancer genome atlas (TCGA); SARC; LIHC; CHOL; KICH; ACC; KM; STAD; std; SKCM; OV; DLBC; ESCA; KIRC; Graph convolutional neural network; CESC; LGG; TCGA; THYM; READ; BRCA; PAAD; Survival analysis; THCA; BLCA; HNSC; LAML; PI; LUSC; GCNN; KIRP
• ISSN: 1046-2023; 1095-9130; 1095-9130
• DOI: 10.1016/j.ymeth.2021.01.004

•Predicting cancer survival outcomes using Graph Convolutional Neural Networks with TCGA dataset.•Combining clinical data and the GCNN improves cancer prognostic prediction.•Interpreting the GCNN model and identifying significant genes with network modules identified by HotNet2. The survival rate of cancer has increased significantly during the past two decades for breast, prostate, testicular, and colon cancer, while the brain and pancreatic cancers have a much lower median survival rate that has not improved much over the last forty years. This has imposed the challenge of finding gene markers for early cancer detection and treatment strategies. Different methods including regression-based Cox-PH, artificial neural networks, and recently deep learning algorithms have been proposed to predict the survival rate for cancers. We established in this work a novel graph convolution neural network (GCNN) approach called Surv_GCNN to predict the survival rate for 13 different cancer types using the TCGA dataset. For each cancer type, 6 Surv_GCNN models with graphs generated by correlation analysis, GeneMania database, and correlation + GeneMania were trained with and without clinical data to predict the risk score (RS). The performance of the 6 Surv_GCNN models was compared with two other existing models, Cox-PH and Cox-nnet. The results showed that Cox-PH has the worst performance among 8 tested models across the 13 cancer types while Surv_GCNN models with clinical data reported the best overall performance, outperforming other competing models in 7 out of 13 cancer types including BLCA, BRCA, COAD, LUSC, SARC, STAD, and UCEC. A novel network-based interpretation of Surv_GCNN was also proposed to identify potential gene markers for breast cancer. The signatures learned by the nodes in the hidden layer of Surv_GCNN were identified and were linked to potential gene markers by network modularization. The identified gene markers for breast cancer have been compared to a total of 213 gene markers from three widely cited lists for breast cancer survival analysis. About 57% of gene markers obtained by Surv_GCNN with correlation + GeneMania graph either overlap or directly interact with the 213 genes, confirming the effectiveness of the identified markers by Surv_GCNN.