Figure 4From: Inferring gene function from evolutionary change in signatures of translation efficiencyMechanism of activity of the putative oxidative stress protection genes. (A) Overview of outcomes of seven experimental assays (columns) performed with the wild-type Escherichia coli and 15 deletion mutants. A larger value denotes a stronger observed effect; values are adjusted so that 0 signifies no effect and 1 signifies strong effect (values <0 and >1 are possible; for details of normalization for each assay, see Additional file 1). DHR-123 and CellROX are fluorescent dyes that measure reactive oxygen species. 2,2′-dipyridyl is an iron chelator. Genes are clustered based on similarity of the normalized response profiles of the mutants across the assays. Dashed lines denote the median. (B, C) A detailed display of the non-normalized measurements of: (B) iron levels and survival in the dipyridyl rescue experiment, or (C) NADPH levels and survival in the corresponding rescue experiment. Data are shown for wild-type E. coli, for lon and recA mutants (well-investigated genes expected not to act by the examined mechanisms), and those candidate genes in which our experiments support the proposed mechanism of action. Error bars show the 95% CI of the mean, determined over at least four replicates.Back to article page