Figure 6.

ROC curve for protein degree as a drug target predictor. Plot of False Positive Rate versus True Positive Rate for a degree rank of the full PIN and five subsets considered as containing higher-confidence interactions: non-predicted interactions include all interactions except those coming from orthologous transfer; LTP includes interactions with an lpr score < 22; MI-IntAct includes interactions with MI-IntAct scores > 0.6; MI-PSICQUIC includes interactions with MI-PSICQUIC scores > 0.7; and B includes the true binary interactions (i.e., potential spoke-represented n-ary data is removed). Theoretically perfect and random classifiers are shown in grey for reference (AUC = 1 and 0.5 respectively).

Mora and Donaldson BMC Bioinformatics 2012 13:294   doi:10.1186/1471-2105-13-294
Download authors' original image