Table 2. Association of expression modules with OS/DFS in the training and validation dataset.

Modules	Gene count	Training dataset (n = 300)						Validation dataset (n = 248)
		OS			DFS			OS
		HR	95% CI	p-value	HR	95% CI	p-value	HR	95% CI	p-value
MEblack	43	0.633	0.458–0.876	0.006**	0.695	0.503–0.961	0.028*	1.486	1.04–2.121	0.029*
MEblue	404	0.992	0.721–1.363	0.959	0.991	0.721–1.362	0.955	1.158	0.809–1.657	0.422
MEbrown	342	1.586	1.149–2.189	0.005**	1.544	1.119–2.131	0.008**	1.664	1.155–2.398	0.006**
MEgreen	113	1.691	1.222–2.34	0.002**	1.609	1.163–2.226	0.004**	1.105	0.774–1.578	0.583
MEred	95	0.756	0.548–1.042	0.088	0.791	0.574–1.09	0.152	1.2	0.84–1.713	0.317
MEturquoise	1,059	1.75	1.264–2.424	0.001***	1.66	1.198–2.299	0.002**	1.208	0.843–1.729	0.303
MEyellow	236	0.654	0.474–0.903	0.010**	0.749	0.543–1.034	0.079	1.201	0.842–1.713	0.313

Notes.

^*p ≤ 0.05.

^**p ≤ 0.01.

^***p ≤ 0.001.

Overall survival (OS), Disease-free survival (DFS). Hazard ratios (HRs), 95% confidence intervals (CI), and p-values were calculated using Cox proportional hazards regression analysis after grouped the gastric cancer patients by the median of gene level.