Skip to main content
. 2016 Nov 28;45(2):556–566. doi: 10.1093/nar/gkw1085

Figure 2.

Figure 2.

Comparison of the recall (primary y-axis) and relative recall improvement (RRI, secondary y-axis, log-scale) of single-sequence target gene predictors and their corresponding MirAncesTar predictors. (A–C) Average (over 100 miRNAs) of the recall (percentage of known targets recovered) as a function of the number of sites being predicted (k). (A) miRanda; (B) RNAhybrid; (C) mirMark with and without PhastCons. (D–F) Recall (at k = 1000 predictions), for each of the 100 miRNAs, for each SSTSP (x-axis) and its corresponding MirAncesTar predictor (y-axis). MiRNAs for which the difference between the two recall values is statistically significant (P-value < 0.05 based on two-tailed Student's t-test) are shown in black.