Figure 5.

Evaluation of cell classification as described in [2]. The accuracy of every $d$-model, i.e. $d$-factor combination, is assessed in terms of classification error (CE), cross-validation consistency ($CVC$) and prediction error (PE). Among all $d$-models the single model with lowest average CE is selected, yielding a set of best models for each $d$. Among these best models the one minimizing the average PE is selected as final model. To determine statistical significance, the observed $CVC$ is compared to the empirical distribution of $CVC$ under the null hypothesis of no interaction derived by random permutations of the phenotypes.