Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: Metaprotein expression modeling for label-free quantitative proteomics

Figure 4

Simulation ROC curves. Receiver operating curves generated from 200 draws from the MCMC chain (red) compared to ROC’s generated from predictors of randomly generated outcomes (blue). The random predictors were generated from a mean zero multivariate normal distribution using a covariance matrix generated from the factor matrix Λ for the corresponding MCMC draw. The outcome variable (response/non-response) was then permuted before the ‘random predictor’ model was fit.

Back to article page