FIGURE 4.

DRUID can rescue recalcitrant data sets. (A) Effect of poorly behaving data sets on transcript half-lives calculated with exogenous spike-ins. A scatterplot comparing the half-lives in HEK293 cells calculated in one biological replicate and a recalcitrant data set with poorly behaving spike-ins. The red dashed line represents the x = y line. (B) Comparison of transcript half-lives calculated with DRUID. As in A, except half-lives were calculated using DRUID. (C) Comparison of half-lives calculated with DRUID. As in B, except for a second, recalcitrant data set, whose spike-ins behaved so poorly that no half-lives could be calculated using exogenous spike-ins. (D) Comparison of yeast half-lives calculated using exogenous spike-ins, otherwise as in A. (E) Comparisons of yeast half-lives calculated using DRUID, otherwise as in A. See also Supplemental Figure S2.