Supervised Wavelet Method to Predict Patient Survival from Gene Expression Data

Table 2

Performance of different Cox models for DLBCL dataset.

# Gene	Method	C index ± se	CPE ± se	±  se	LR ± se	IBS ± se
40	Supervised wavelet	0.755 ± 0.005	0.744 ± 0.004	0.401 ± 0.011	78.739 ± 1.815	0.237 ± 0.007
	Supervised PCA	0.711 ± 0.004	0.695 ± 0.003	0.270 ± 0.000	42.636 ± 1.762	0.245 ± 0.005
	Supervised PLS	0.723 ± 0.003	0.698 ± 0.003	0.294 ± 0.007	55.883 ± 1.449	0.250 ± 0.005
30	Supervised wavelet	0.723 ± 0.005	0.727 ± 0.007	0.325 ± 0.013	70.303 ± 2.618	0.244 ± 0.004
	Supervised PCA	0.709 ± 0.004	0.692 ± 0.003	0.262 ± 0.008	42.087 ± 1.825	0.245 ± 0.003
	Supervised PLS	0.713 ± 0.002	0.697 ± 0.002	0.289 ± 0.007	54.898 ± 1.418	0.251 ± 0.004
20	Supervised wavelet	0.730 ± 0.002	0.714 ± 0.002	0.323 ± 0.009	59.708 ± 2.699	0.243 ± 0.004
	Supervised PCA	0.709 ± 0.003	0.688 ± 0.003	0.260 ± 0.008	41.327 ± 2.079	0.245 ± 0.003
	Supervised PLS	0.719 ± 0.002	0.696 ± 0.003	0.282 ± 0.006	53.130 ± 1.486	0.249 ± 0.004
10	Supervised wavelet	0.703 ± 0.004	0.686 ± 0.005	0.255 ± 0.007	49.838 ± 1.832	0.248 ± 0.003
	Supervised PCA	0.699 ± 0.005	0.686 ± 0.003	0.254 ± 0.013	41.056 ± 2.045	0.252 ± 0.004
	Supervised PLS	0.701 ± 0.003	0.684 ± 0.003	0.255 ± 0.007	45.648 ± 2.241	0.254 ± 0.006