Evaluation of the different pSDC metrics and RNA structure prediction algorithm’s performances. A.) Receiver Operator Characteristic (ROC) analysis of three representative metric/algorithm combinations indicating significant differences in the predictive performance. The three curves are for the RNAsubopt prediction of the Z centroid structure (for the first 10,000 structures), base-pair probability as computed from the partition function of RNAfold, and the MFE structures predicted by sFold. B.) Summary of all Area Under the Curve values for ROC analysis of metric/algorithm combinations. In general, most algorithms perform equivalently for identifying the mutations that disrupt RNA structure.
Ritz et al. BMC Genomics 2012 13(Suppl 4):S6 doi:10.1186/1471-2164-13-S4-S6