Whole data analysis | 200 CVs | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|

Set | Model | Factors Included | Variance (90% posterior confidence region) | Log likelihood^{e} | Effective number of parameters (pD) | Deviance information criteria (DIC) | Average CV-AUC (SD)^{f} | Proportion of times model in column had AUC > model in the row | ||||

Covariates^{a} | CNV^{b} | METH^{c} | miRNA^{d} | M13, M16, M19: omic only | M14, M17, M20: COV + omic | |||||||

Set 1 (n = 270) | M12: COV | X | — | −125.5 | 8.1 | 259.0 | 0.699 (0.009)^{g} | <0.01 | >0.99 | |||

M13: CNV | X | 0.637 (0.155; 1.124) | −112.1 | 26.8 | 250.9 | 0.653 (0.012)^{h} | — | >0.99 | ||||

M14: COV + CNV | X | X | 0.398 (0.070; 0.736) | −110.5 | 24.5 | 245.6 | 0.714 (0.009)^{i} | — | — | |||

Set 2 (n = 199) | M15: COV | X | — | −88.7 | 8.4 | 185.7 | 0.667 (0.013)^{g} | 0.60 | >0.99 | |||

M16: METH | X | 0.652 (0.086; 1.261) | −76.6 | 18.7 | 171.8 | 0.672^{g}^{,} (0.017)^{h} | — | 0.76 | ||||

M17: COV + METH | X | X | 0.402 (0.032; 0.739) | −78.9 | 18.5 | 176.3 | 0.684 (0.013)^{h} | — | — | |||

Set 3 (n = 167) | M18: COV | X | — | −71.2 | 8.2 | 150.6 | 0.747 (0.011)^{g} | <0.01 | 0.29 | |||

M19: miRNA | X | 0.338 (0.072; 0.615) | −75.2 | 13.5 | 163.8 | 0.623 (0.018)^{h} | — | >0.99 | ||||

M20: COV + miRNA | X | X | 0.179 (0.029; 0.324) | −67.3 | 13.8 | 148.5 | 0.744 (0.011)^{g} | — | — |

a Age: African American, Y/N; lobular (Y/N); cancer subtype and stage.

b Copy-number variants.

c Methylation.

d Whole-genome RNA-seq.

e Estimated posterior mean of the log likelihood.

f Average over 200 tenfold CVs.

g,h,i The same letter indicates that the models are no different (empirical

*P*< 0.05).