Correction for Sharon et al., Commensal bacteria play a role in mating preference of Drosophila melanogaster

MICROBIOLOGY Correction for “Commensal bacteria play a role in mating preference of Drosophila melanogaster,” by Gil Sharon, Daniel Segal, John M. Ringo, Abraham Hefetz, Ilana Zilber-Rosenberg, and Eugene Rosenberg, which appeared in issue 46, November 16, 2010, of Proc Natl Acad Sci USA (107:20051–20056; first published November 1, 2010; 10.1073/pnas.1009906107).

The authors note the following: “The mating frequencies reported in Table 1 of this paper did not follow a multinomial distribution, making the statistical analysis inapplicable. This problem was obviated by considering only the first matings in each experimental unit and computing odds ratios. After submitting the paper, we continued to perform experiments identical in design to those we reported. In the table below, we combined the results of those additional replicate experiments with those already reported. From the new analysis, we now find that experiment 4, in which flies were infected with a mixture of Lactobacillus spp., assortative mating was not restored. Otherwise, the conclusions of the article were not changed by our reanalysis. We acknowledge the statistical advice of Dan Yekutieli and thank Tal Lahav for calculating the odds ratios and their 95% confidence intervals, and for performing the chi-squared tests presented in the corrected Table 1.”

Table 1.

The role of bacteria in diet-induced mating preference of D. melanogaster

Experiment	Fly treatment*	N†	OR‡	95% CI	P value‡
1	Starch-grown × CMY-grown	18	3.21	2.14–4.81	1.8 × 10−8
2	Experiment 1 after antibiotics	11	1.04	0.63–1.71	0.9888
3	Experiment 2 after infection of starch-grown flies with homologous bacteria§	6	2.68	1.40–5.11	0.0477
4	Experiment 3 with Lactobacillus spp. replacing homologous bacteria	4	1.76	0.74–4.19	0.2912
5	Experiment 3 with Lactobacillus plantarum replacing homologous bacteria	7	2.14	1.35–3.39	0.0019
6	Infection control (no added bacteria)	4	1.26	0.53–3.00	0.7712

^*After all treatments, the flies were grown for one generation in CMY medium before performing the mating preference test.

^†N is the number of replicate experiments.

^‡Cochran-Mantel-Haenszel Odds Ratio and P value are from the Cochran-Mantel-Haenszel Chi-squared test (34, 35).

^§Antibiotic-treated starch- and CMY-grown flies were infected with bacteria isolated from their respective growth medium (before antibiotic treatment).

The corrected Table 1 appears below.