Oviposition in Drosophila is rhythmic when registered with our semiautomated egg collection device.

A, C, G, I: Average eggs collected as a function of time in four different experiment (Canton-S, LD: A, Canton-S, DD:C; YW: G; perS: I). The average was made with all flies (Strongly rhythmic, weakly rhythmic and arrhythmic). White and dark grey bars represent periods of lights on and off respectively (LD), whereas light gray bars represent subjective days, DD, (i.e times where lights were on at rearing, but are now off). B, D, H, J: Lomb-Scargle periodograms of all genotypes, made with all flies, strongly rhythmic, weakly rhythmic and arrhythmic. Red and green horizontal lines represent significances of 0.05 and 0.01, respectively. E, K: Percentage of females with rhythmic oviposition (E: Canton-S in LD and DD, K: YW and perS). The rhythmic flies include both strongly and weakly rhythmic flies. F, L: Period of oviposition rhythms for rhythmic individual flies (F: Canton-S in LD and DD, L: YW and perS). ns: non significant, ***p<0.001 (chi-squared test for scatter plots and proportions).

Circadian rhythmicity of oviposition is dramatically reduced when the molecular clock is disrupted in all clock neurons, but not when only LNv or DN1 neurons are affected.

A, F, K, P: Schematic diagram of the neurons (painted in red) where the molecular clock has been disrupted. B, D, G, I, L, N, Q, S: Average eggs collected as a function of time. The average was made with all flies (strongly rhythmic, weakly rhythmic and arrhythmic). White and dark grey bars represent periods of lights on and off respectively (LD), whereas light gray bars represent subjective days, DD, (i.e times where lights were on at rearing, but are now off).C, E, H, J, M, O, R, T: Lomb-Scargle periodograms of all genotypes (made with all flies, rhythmic, weakly rhythmic and arrhythmic). Red and green horizontal lines represent significances of 0.05 and 0.01, respectively. B, C: +>UAS-perRNAi, n=22. D, E: Clk856-Gal4>UAS-perRNAi, n=26. G, H: +>UAS-perRNAi, n=18. I, J: PdfDicer-Gal4>UAS-perRNAi, n=34. L, M:+>UAS-perRNAi, n=38. N, O: Clk4.1-Gal4>UAS-perRNAi, n=40. Q, R: +>UAS-kir2.1, n=36. S, T:Clk4.1-Gal4>UAS-kir2.1, n=35.

Disruption of the molecular clock in Cry+ LNd neurons drastically reduces the circadian rhythmicity of oviposition.

A, H: Schematic diagram of the neurons (painted in red) where the molecular clocks have been disrupted. B, D, F, I, K, M: Average eggs collected as a function of time. The average was made with all flies (Strongly rhythmic, weakly rhythmic and arrhythmic). White and dark grey bars represent periods of lights on and off respectively (LD), whereas light gray bars represent subjective days, DD, (i.e times where lights were on at rearing, but are now off).C, E, G, J, L, N: Lomb-Scargle periodograms of all genotypes (made with all flies, rhythmic, weakly rhythmic and arrhythmic). Red and green horizontal lines represent significances of 0.05 and 0.01, respectively. B, C: +>UAS-perRNAi, n= 15. D, E: Mai179-Gal4;pdf-Gal80>+, n= 17. F, G: Mai179-Gal4;pdf-Gal80>UAS-perRNAi, n=28. I, J: +>UAS-cycDN, n=15. K, L: MB122Bsplit-Gal4>+, n=14. M, N: MB122Bsplit-Gal4>UAS-cycDN, n=30.

Disruption of the molecular clock in Cry+ LNd neurons does not alter the circadian rhythmicity of locomotor activity of mated female flies.

A: Average activity recording during 3 days in LD and 7 in DD. Light grey bars represent subjective days. B: Percent of rhythmic flies in DD. C: Periods of locomotor activity in DD of individually rhythmic flies. Mean ± SEM indicated. Each dot represents one fly. ns: non significant, *: p<0.05 (chi-squared test for the comparison of proportions, and one-way-ANOVA with post-hoc Tukey for the scatter plots.

Direct synaptic connections between circadian clock neurons and oviposition-related neurons in the hemibrain dataset.

A: Schematic diagrams showing the different neuron clusters analyzed (in color). B: Connection between Cry+ LNd and oviIN neurons. C: Connection between Cry-LNd 2and pC1b neurons. D: Network representation of the connectivity between Cry+ LNd and oviIN neurons. E: Network representation of the connectivity between Cry- LNd and pC1b neurons. Numbers give the number of synaptic contacts, which represent the strength of the connections. Only intermediate (between 3 and 9 synaptic contacts) and strong (>9 synaptic contacts) connections are considered.

Scheme of connections between circadian clock and oviposition-related neuron clusters of the same hemisphere in the hemibrain dataset.

Each circle represents a neuron cluster, comprising different numbers of neurons (some clusters comprise only one neuron). Clusters and connections involved in the connectivity between circadian clock and oviposition sets have been colored. In each connection the arrow points to the post-synaptic cluster. OviIN neurons are bidirectionally connected to every neuron of the clusters inside the square.

All fly strains used in this study.

Semi-automated egg collection device.

A: View from above. The white square is the wooden base to which there are two bars attached that hold the mechanical arm (in black), moved by a stepper motor (black box in the upper right corner of the wooden base). The arm shifts forward (downward in the picture) each fly chamber (light green box) from one well to the next. B: Detail of the tracks (dark green) holding the fly chambers, and the food wells filled with banana medium (brown) and a drop of yeast (white). C: Lateral view of the tracks. D: Detail from above, showing the plastic piece used to press the chambers against the tracks.

Disruption of the molecular clock in PDF+ neurons causes a shortening of the egg-laying period at the individual level.

Period of egg-laying rhythms for rhythmic individual flies (UAS-perRNAi, n=8 (out of 18), PdfDicer-Gal4>UAS-perRNAi, n=16 (out of 34). ***: p < 0.001 (chi-squared test).

Disruption of the molecular clock in Cry+ LNd neurons drastically reduces the circadian rhythmicity of oviposition.

A: Schematic diagram of the neurons (painted in red) where the molecular clocks have been disrupted. B, D: Average eggs collected as a function of time. The average was made with all flies (Strongly rhythmic, weakly rhythmic and arrhythmic). White and dark gray bars represent periods of lights on and off respectively (LD), whereas light gray bars represent subjective days in DD, (i.e times where lights were on at rearing, but are now off). C, E: Periodograms of the average time series of the number of eggs laid by all females. F, G: Periodograms of the individual time series of the number of eggs laid. Red and green horizontal lines represent significances of 0.05 and 0.01, respectively. B, C, F: control (+>UAS-perRNAi, n=25). D, E, G: MB122BsplitGal4>perRNAi (MB122B-splitGal4>UAS-perRNAi, n=20).

Individual periodograms for the records obtained after applying detrending for the experiment in DD with Mai179-Gal4;pdf-Gal80 >UAS-perRNAi (n=28) female flies described in the main text.

Green horizontal lines represent significances of p=0.05.

Individual periodograms for the records obtained after applying detrending for the experiment in DD with Mai179-Gal4;pdf-Gal80>+ (n=17) female flies described in the main text.

Green horizontal lines represent significances of p=0.05.

Individual periodograms for the records obtained after applying detrending for the experiment in DD with +>UAS-perRNAi, (n=15) female flies described in the main text.

Green horizontal lines represent significances of p=0.05.

Individual periodograms for the records obtained after applying detrending for the experiment in DD MB122Bsplit-Gal4>UAS-cycDN (n=30) female flies described in the main text.

Green horizontal lines represent significances of p=0.05.

Individual periodograms for the records obtained after applying detrending for the experiment in DD with MB122Bsplit-Gal4>+ (n=14) female flies described in the main text.

Green horizontal lines represent significances of p=0.05.

Individual periodograms for the records obtained after applying detrending for the experiment in DD with +>UAS-cycDN (n=15) female flies described in the main text.

Green horizontal lines represent significances of p=0.05.

Direct synaptic connections between LPN and oviposition-related neurons in the hemibrain dataset.

A: Scheme showing the neurons analyzed (in color). B: Connection between LPN and oviIN neurons. C: Network representation of the connectivity between these neurons. Numbers give the number of synaptic contacts, which represent the strength of the connections. Only intermediate (between 3 and 9 synaptic contacts) and strong (> 9 synaptic contacts) connections are considered

Downregulation of per in all clock brain neurons impairs locomotor circadian activity:

Analysis of the period (τ) and percentage of rhythmicity (R) between genotypes. Different letters indicate statistically significant differences (p < 0.05, One-way ANOVA with a Tukey post-hoc test). n indicates the number of animals tested.

Effect of noise in the assessment of rhythmicity of periodic signals.

A,C: Periodic signal with Gaussian noise (average=0, sd=2), sampled at intervals of 30 minutes (A) or 4 hours (C). B,D: Periodograms of the signals shown in panel A and C, respectively. E: Comparison of periodogram peak heights (relative to the significance level) for 150 signals sampled at different intervals. pp is the power of the periodogram peak, whereas pσ denotes the power corresponding to a significance level σ = 0.05. The green line, corresponding to ps = pσ, separates experiments where the peaks are significant (above) from those where peaks are not significant (below).

Generation and decomposition of a decaying noisy rhythmic synthetic signal.

A: Modulating decay function. B: Noisy rhythmic signal. D: Signal given by the product of signals A and B. F: Estimate of the decay function. G: Estimate of the rhythmic signal. C, E, H: Periodogram analysis of the signals in B, D, G, respectively. Green horizontal lines mark the power corresponding to a significance level of 0.05. Dashed lines are only guides to the eye.

Period determination for different synthetic signals.

Periodograms of rhythmic noisy signals with period T (green triangles), of the same signal with decay applied (blue squares), and of the estimation of the signal (red circles). A: T = 20. B: T = 24. C: T = 28. D: α = 0 (i.e. purely arrhythmic signal). Each horizontal line corresponds to the significance level of 0.05 for the periodogram with matching color. Dashed lines are only guides to the eye.

Accuracy of period determination for different synthetic signals.

A: Period Tnestimated for 2000 noisy rhythmic signals generated with T between 19 and 28 hours. B: Period Tnd estimated for the same signals as in A when decay is applied, as a function of the period estimate Tnof the original signal. C: Period Tndw estimated when our method is applied to the signals in B, as a function of the period estimate Tn of the original signal. Horizontal and vertical dashed lines correspond to the values present in the corresponding period grid. In B and C, the size of the symbols is proportional to the number of signals with the periods indicated by the crossing grid lines.

Presence of false negatives and absence of false positives.

P-values obtained for 100 rhythmic (A) and 100 arrhythmic (B) synthetic signals. Left: Noisy rhythmic signals. Center: Signals obtained by applying decay to the previous signals. Right: Signals obtained by applying our method to the previous ones. The green dashed line indicates the level of significance used (0.05).

Individual egg records for the first synthetic experiment with 28 flies.

Records obtained after applying detrending for the first synthetic experiment de-scribed in the text.

Periodograms for the records obtained after applying detrending for the first syn-thetic experiment described in the text.

Results for the first synthetic experiment described in the text.

A: Population periodogram. B: Individual periods (Left: T of the underlying rhythm, center: period estimate of the signal without decay, right: estimated periods of the records with a significant peak in the periodogram).

Individual egg records of the second synthetic experiment with 28 flies.

Records obtained after applying detrending for the second synthetic experiment described in the text.

Periodograms for the records obtained after applying detrending for the second synthetic experiment described in the text.

A few rhythmic flies do not make a population rhythmic.

A: Population peri-odogram for the second experiment mentioned in the text. B: Population periodogram for the first experiment without the five most rhythmic flies. C: Population periodogram for the second experiment with 5 flies replaced by the 5 rhythmic flies of the first experiment.

Individual egg records of an experiment in DD with 18 CantonS female flies.

Records obtained after applying detrending for the experiment in DD with 18 CantonS female flies described in the text.

Periodograms for the records obtained after applying detrending for the experiment in DD with 18 CantonS female flies described in the text.

Individual egg records of an experiment in DD with 19 per01 female flies.

Records obtained after applying detrending for an experiment in DD with 19 per01 female flies.

Periodograms for the records obtained after applying detrending for an experiment in DD with 19 per01 female flies.

Persistence of population rhythmicity with or without individually rhythmic flies.

A: Population periodogram for the experiment with 18 CantonS female flies mentioned in the text. B: Population periodogram for the same experiment as in A, but without 8 individually rhythmic flies. C: Population periodogram for the experiment with 19 per01 female flies mentioned in the text