Identification of a Robust Five-Gene Risk Model in Prostate Cancer: A Robust Likelihood-Based Survival Analysis

Table 2

The best prognosis-related model results selected by “Rbsurv” package in R.

Order	Gene	nloglik	AICs	Selected
0	0	275.87	551.73
1	TACR2	261.70	525.41
2	FAM72D	255.02	514.03
3	PLEK2	251.31	508.62
4	FA2H	248.07	504.15
5	ARHGAP33	243.38	496.76
6	TRIM74	242.55	497.11
7	TRIM73	242.03	498.05
8	SCNN1D	242.01	500.02
9	KRTAP5-1	241.66	501.32
10	CCDC180	240.98	501.97
11	MXD3	240.97	503.94
12	GPC2	239.75	503.5
13	SSPO	239.74	505.48
14	CPLX1	239.59	507.19
15	AL157935.2	237.49	504.99
16	SOX8	236.62	505.24
17	FGF17	233.77	501.53
18	SPRED3	233.21	502.42
19	SEC61A2	232.82	503.63

AIC: Akaike information criterion score; nloglik: negative log-likelihood.