(A, B) Representative double-plotted actograms for 5- to 10-day-old male flies maintained on standard food (A) or food supplemented with 2 mm NAA (B) for 3 days in 12 hr light:12 hr dark (12:12 LD) and 7 days in DD. Left: parental control UAS-Kir2.1, centre: parental control PDF-GAL4; AGES, right: experimental PDF-GAL4/UAS-Kir2.1; AGES/+. Bars indicate LD cycle, grey shaded days indicate constant darkness. (C) Amplitude of circadian rest:activity rhythms on DD days 2–8 represented by fast Fourier transform (FFT) power at 24 hr for male PDF-GAL4;AGES > UAS-Kir2.1 flies and their parental controls on standard food (orange) and food supplemented with 2 mM NAA (blue). Points represent individual flies, box shows 25–75% confidence interval, median line, and outliers. (D) Twenty-four hr FFT power as in (C) for female PDF-GAL4;AGES > UAS-Kir2.1 flies and their parental controls. (E) Twenty-four hr FFT power for as in (C) male PDF-GAL4-Geneswitch > UAS-Kir2.1 flies and their parental controls maintained on vehicle control food (orange) and food supplemented with 466 mM RU-486 (red). (F) Twenty-four hr FFT power as in (E) for female PDF-GAL4-Geneswitch > UAS-Kir2.1 flies and their parental controls. For all panels, means were compared by two-way ANOVA by genotype and food substrate, see Figure 6—source data 1 for raw data, p-values, and key resource data. Means sharing the same letter are not significantly different from one another by Tukey’s post hoc test (p > 0.05).