Fig. 3
From: Detecting and correcting the binding-affinity bias in ChIP-seq data using inter-species information
Comparison of models \({\mathcal {M}_{-}^{\mathrm {C}}}\) and \({\mathcal {M}_{\text {BA}}^{\mathrm {C}}}\). a Mean classification performance and standard error of the two models \({\mathcal {M}_{-}^{\mathrm {C}}}\) and \({\mathcal {M}_{\text {BA}}^{\mathrm {C}}}\) quantified by the area under the receiver operating characteristic curve. We find for each of the five data sets a significantly increased classification performance for \({\mathcal {M}_{\text {BA}}^{\mathrm {C}}}\) compared to \({\mathcal {M}_{-}^{\mathrm {C}}}\). Examples for ROC curves are shown in Additional file 1: Figure S10, Figure S11, Figure S12, Figure S13, Figure S14 and Figure S15. b Mean information content and standard error of the motifs predicted by the two models \({\mathcal {M}_{-}^{\mathrm {C}}}\) and \({\mathcal {M}_{\text {BA}}^{\mathrm {C}}}\). We find for each of the five data sets a significantly decreased information content in motifs predicted by \({\mathcal {M}_{\text {BA}}^{\mathrm {C}}}\) compared to \({\mathcal {M}_{-}^{\mathrm {C}}}\) (p<4.0×10−18)