Fig. 8

CV values of the relative abundance predictions from different mutants. a Smoothened distributions of CV values of the relative abundance predictions generated by single mutants (blue curve), double mutants (green curve), and triple mutants (red curve) among the 129 novel mutants. Mean ± standard deviation for each distribution (listed in Additional file 1: Table S12) is depicted by a single horizontal bar. These predictions are generated by the parameter vectors in Ensemble 16. b-g CV values of the relative abundance predictions for the WT strain (in b) and the five most informative (highest ranked based on a prediction variability statistic) novel mutants (Mutant 90 in c, Mutant 21 in d, Mutant 57 in e, Mutant 11 in f, and Mutant 85 in g. These predictions are generated by the parameter vectors in Ensemble 16. h Distributions of CV values of the relative abundance predictions generated by Mutant 21 (a double mutant shown in blue curve), which is a combination of two single mutants: Mutant 2 (green curve) and Mutant 6 (red curve). Mean ± standard deviation for each distribution (listed in Additional file 1: Table S13) is depicted by a single horizontal bar. These predictions are generated by the parameter vectors in Ensemble 16