Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: Bayesian inference for identifying tumour-specific cancer dependencies through integration of ex-vivo drug response assays and drug-protein profiling

Fig. 2

Prediction performance: Each point in each figure corresponds to a model fitted under a given value of hyper-parameter using either the original, imputed \(\varvec{Y}_{\textrm{imp}}\) or the incomplete, logit-(log-)transformed \(\varvec{Y}_{\textrm{logit}}\)(\(\varvec{Y}_{\textrm{log}}\)). The horizontal coordinate of the point is the normalized MSE of the model estimated using 3-fold CV, and the vertical coordinate is the Intersection-over-Union score between the subset of kinases selected by the fitted model and the corresponding subset of kinases selected by DepInfeR. The vertical dashed and solid lines correspond to the estimated normalized MSE of multivariate Random Forest and DepInfeR based on the original dataset \(\{\varvec{X}_{\textrm{imp}}, \varvec{Y}_{\textrm{imp}}\}\)

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us