For the article by Y. D. Kelkar, D. S. Phillips, and H. Ochman (G3 Genes|Genomes|Genetics 5: 1247–1252) entitled “Effects of Genic Base Composition on Growth Rate in G+C-Rich Genomes,” the original publication requires a correction that impacts one of the conclusions. One of the Caulobacter crescentus constructs (gfp156) has a high doubling rate in both induced and noninduced conditions, and the authors are grateful to Dr. Bert Ely for calling this to their attention.
This feature affects the reported relationship between G+C contents and C. crescentus growth rates, and after removal of this data point from analyses, the regression no longer reaches significance in either induced or noninduced conditions (p values increase to 0.18 and 0.46, respectively). In light of these results, these experiments cannot currently be taken as evidence of selection on base composition in C. crescentus. Experiments that include additional constructs in the low G+C range would be required before making any further conclusions. It should be noted that values reported for gfp156 are accurate and reproducible but for the sake of statistical diligence, they cannot be included in the regression analysis. The authors’ conclusions for Pseudomonas aeruginosa, whose results fit the inferred trend, remain the same.