Research Article
A Risk Stratification Model for Lung Cancer Based on Gene Coexpression Network and Deep Learning
Figure 2
Selection of representative genes of survival-related network modules. (a) To construct risk stratification model, representative genes were selected according to the gene module membership. Gene module membership was correlated with the significance of association between individual gene expression and survival. -axis represents statistical significance calculated by univariate Cox analysis of individual genes. A strong correlation was found in the red and turquoise modules ( and < 1 × 10−19 for red module; and < 1 × 10−23 for turquoise module). Coexpression networks of red (b) and turquoise (c) modules were visualized. Note that 160 genes among 880 genes of turquoise module and their connections were shown. 160 genes were selected according to the gene module membership. Size of nodes is proportional to gene module membership.
| (a) |
| (b) |
| (c) |