Table 2.
Nine candidate genes of miR-181 that are significantly associated with poor OS
| Gene | Training sets (n = 183) |
Validation set (n = 271) |
||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Coefficient average | FDR | USA-set-II-35S (n = 35) |
USA-set-III (n = 87) |
Germany-set-I (n = 61) |
Netherlands-set-I (n = 271) |
|||||||||
| HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | |||
| HOXA7 | 0.477 | 0.03 | 2.09 | 1.36-3.19 | < .001 | 1.22 | 0.80-1.85 | .36 | 1.65 | 1.15-2.36 | .01 | 1.34 | 1.14-1.57 | < .001 |
| HOXA9 | 0.483 | 0.02 | 2.12 | 1.38-3.25 | < .001 | 1.18 | 0.79-1.76 | .42 | 1.70 | 1.16-2.47 | .01 | 1.41 | 1.21-1.64 | < .001 |
| HOXA11 | 0.433 | 0.03 | 1.47 | 1.06-2.04 | .02 | 1.70 | 1.18-2.43 | .004 | 1.47 | 0.83-2.59 | .18 | 1.00 | 0.85-1.17 | .99 |
| PBX3 | 0.546 | 0.02 | 2.06 | 1.42-2.97 | < .001 | 1.61 | 1.19-2.18 | .002 | 1.55 | 1.06-2.27 | .02 | 1.17 | 1.00-1.36 | .04 |
| MRPS18B | 0.407 | 0.03 | 1.53 | 1.06-2.23 | .02 | 1.74 | 1.27-2.39 | < .001 | 1.27 | 0.81-2.00 | .29 | 1.04 | 0.89-1.22 | .59 |
| PLA2G4A | 0.531 | < .001 | 1.79 | 1.21-2.64 | .003 | 1.60 | 1.16-2.22 | .005 | 1.72 | 1.11-2.65 | .01 | 1.31 | 1.11-1.54 | .002 |
| TRPS1 | 0.395 | 0.05 | 1.53 | 1.08-2.18 | .02 | 1.41 | 1.02-1.95 | .04 | 1.52 | 1.02-2.25 | .04 | 1.37 | 1.18-1.59 | < .001 |
| GNS | 0.258 | < .001 | 1.71 | 1.02-2.88 | .04 | 1.69 | 1.21-2.36 | .002 | 0.75 | 0.49-1.15 | .18 | 1.02 | 0.85-1.21 | .87 |
| PGPEP1 | 0.488 | 0.03 | 2.68 | 1.38-5.22 | .004 | 1.30 | 0.93-1.80 | .12 | 1.25 | 0.87-1.78 | .23 | 1.14 | 0.99-1.32 | .07 |
Coefficient average indicates the average correlation coefficient from the 3 training sets.