Code and metadata for Quentin Geissmann*, Esteban J. Beckwith* and Giorgio F. Gilestro, 2019

Figure layout and reference for the manuscript

Fig 1, baseline sleep/quiescence

• summary : Undisturbed baseline behaviour for four full days males vs females.
• experiment -- 20160404_overnight_dsd
• script -- baseline.R
• Ns: N_M = 485, N_F = 881
• figure material:
  1. overnight_dsd_baseline.pdf, page 1 -- sleep fraction.
  2. overnight_dsd_baseline.pdf, page 2 -- quiescence (i.e. not moving) fraction.
  3. overnight_dsd_baseline.pdf, page 3 -- like 1., but wrapped over 24h
  4. overnight_dsd_baseline.pdf, page 4 -- like 2., but wrapped over 24h
  5. sorted_baseline_overview.pdf, pages 1 & 2 -- distribution of individual sleep in females
  6. sorted_baseline_overview.pdf, pages 3 & 4 -- distribution of individual sleep in males
  7. sorted_baseline_overview.pdf, pages 5 & 6 -- distribution of individual quiescence in females
  8. sorted_baseline_overview.pdf, pages 7 & 8 -- distribution of individual quiescence in males
  9. sorted_baseline_overview.svg example of how to put figures marginal distribution and overview together
  10. sorted_baseline_overview.pdf, page 9 -- like page 2. but showing the 19 females for which we have a video
  11. sorted_baseline_overview.pdf, page 10 -- the 19 picked females, as an overview plot. The names on the y axis indicates where to find the video. It reads <experiment_id>|<region_id> -> <video_id>|<region_id>

Fig 2, baseline ternary plots

• summary : same data as fig 1. Introducing the triangle.
• experiment -- 20160404_overnight_dsd
• script -- baseline.R
• Ns are as above
• figure material:
  1. overnight_dsd_baseline.pdf, page 5 -- individual fine grain behaviour for the 19 females (see fig 1), over 48h day, 1min resolution (same data as used in triangle, linking figure 1 and 2).
  2. ternary_plot_video.pdf, all pages -- location in behavioural space for each animal (average across days to wrap on circadian time) 15min /page. Suggested conversion to gif: convert -delay 20 -density 200x200 -loop 0 ternary_plot_video.pdf -background white -alpha remove -alpha off -coalesce -layers optimize ternary_plot.gif
  3. ternary_plot.pdf, page 10 -- summary of 1., wraped in one figure.
  4. ternary_plot.pdf, page 1 and ternary_female_path.pdf, page 1 -- suggested figure legend for 2.
  5. ternary_plot.pdf, page 2-9 -- like 2. but for different population quantiles (based on sleep amount)
  6. ternary_plot.pdf, page 11 -- Overall relative position of animals given their behaviour and sex. For females, micro-mov. is on the food!
  7. sleep_dam_overestimate.pdf page 1 -- relationship between dam-scored and ethoscope-scored sleep (global average, L+D). We also show overall proportion of micromovement as dot size

Fig 3, virgin vs mated females

• summary : Effect of mating on female sleep/quiescence
• experiment -- 20170814_virgin_vs_mated
• script -- script.R
• N_mated > 85, N_virgin > 150
• figure material:
  1. virgin_vs_mated.pdf, page 1 -- sleep fraction.
  2. virgin_vs_mated.pdf, page 2 -- quiescence fraction.
  3. virgin_vs_mated.pdf, page 3 -- position 0 <=> food, 1 <=> cotton wool.
  4. virgin_vs_mated.pdf, page 4 -- like 1., but wrapped over 24h (post-mating data only)
  5. virgin_vs_mated.pdf, page 5 -- like 2., but wrapped over 24h (post-mating data only)
  6. virgin_vs_mated-ternary_plots.pdf, page 2 -- ternary plot post-mating. FALSE and TRUE (i.e. left and right) facets are for virgin and mated, respectively.

Fig 4, Sleep deprivation, 12h overnight.

• summary : Effect of sleep deprivation with 10 different (20 - 1000s) intervals.
• experiment -- 20160404_overnight_dsd
• script -- rebound.R
• N > 45 / treatment * sex (see last page)
• figure material:
  1. overnight_dsd_rebound.pdf, page 1 -- sleep fraction 20s vs control.
  2. overnight_dsd_rebound.pdf, page 2 -- sleep fraction 20s vs control, zoomed in.
  3. overnight_dsd_rebound.pdf, page 3 -- sleep fraction all vs control
  4. overnight_dsd_rebound.pdf, page 4 -- quiescence fraction 20s vs control.
  5. overnight_dsd_rebound.pdf, page 5 -- quiescence fraction 20s vs control, zoomed in.
  6. overnight_dsd_rebound.pdf, page 6 -- quiescence fraction all vs control
  7. overnight_dsd_rebound.pdf, page 7 -- absolute number of stimuli 20s vs control.
  8. overnight_dsd_rebound.pdf, page 8 -- absolute number of stimuli 20s vs control, zoomed in.
  9. overnight_dsd_rebound.pdf, page 9 -- absolute number of stimuli all vs control
  10. overnight_dsd_rebound.pdf, page 10 -- same as 7., but relative to theoritical max num of stimuli
  11. overnight_dsd_rebound.pdf, page 11 -- same as 8., but relative to theoritical max num of stimuli
  12. overnight_dsd_rebound.pdf, page 12 -- same as 9., but relative to theoritical max num of stimuli
  13. overnight_dsd_rebound.pdf, page 15 & 16 -- sleep amount in the 3 (page 15) or 6 hours following SD (ZT 0-[3|6]). Bootstrap error bars
  14. overnight_dsd_rebound.pdf, page 21 -- observed quiescence - predicted quiescence in 3h (in min). Groups are to be compared to red dotted line (see lab meeting 20171102) the expectation. Stats are uncorrected Wilcoxon (unpaired) tests.
  15. overnight_dsd_rebound.pdf, page 22 -- Probability of having a rebound (as defined above) > 0 in 3h (in min). Groups are to be compared to red line (50%), the expectation. Stats are uncorrected Binomial tests.

Fig 5, prolonged SD.

• summary : Effect of 9.5 days sleep deprivation 20s intervals.
• experiment -- 20170209_prolonged_sd
• script -- rebound.R
• N > 90 / treatment * sex (see last page)
• figure material:
  1. prolonged_sd_rebound.pdf, page 1 -- sleep fraction 20s vs control.
  2. prolonged_sd_rebound.pdf, page 2 -- sleep fraction 20s vs control, zoomed in.
  3. prolonged_sd_rebound.pdf, page 3 -- quiescence fraction 20s vs control.
  4. prolonged_sd_rebound.pdf, page 4 -- quiescence fraction 20s vs control, zoomed in.
  5. prolonged_sd_rebound.pdf, page 5 -- absolute number of stimuli 20s vs control.
  6. prolonged_sd_rebound.pdf, page 6 -- absolute number of stimuli 20s vs control, zoomed in.
  7. prolonged_sd_rebound.pdf, page 8 -- relative number of stimuli (max = one each 20s) for the 20s interval group. wrapped over circadiand day, from t = days(1).
  8. prolonged_sd_rebound.pdf, page 9 & 10 -- sleep amount in the 3 (page 15) or 6 hours following SD (ZT 0-[3|6]). Bootstrap error bars.
  9. prolonged_sd_rebound.pdf, page 15 -- observed quiescence - predicted quiescence in 3h (in min). Groups are to be compared to red line, the expectation. Stats are uncorrected Wilcoxon (unpaired) tests. #TODO express in min #TODO add some alpha
  10. prolonged_sd_rebound.pdf, page 16 -- Probability of having a rebound (as defined above) > 0 in 3h (in min). Groups are to be compared to red line (50%), the expectation. Stats are uncorrected Binomial tests.
  11. prolonged_sd_surv.pdf, page 1 -- Survival curve (includes animals that died before rebound, which are no in sleep trace data)

Supplementary figures

Sup fig A: DSD for 4 hours only

• summary : Effect of dynamic sleep deprivation from zt 20 to 24.
• experiment -- 20160722_time_window_dsd
• script -- rebound.R
• N > 50 / treatment * sex (see last page)
• figure material:
  1. time_window_dsd_rebound.pdf, page 1 -- sleep fraction 20s vs control.
  2. time_window_dsd_rebound.pdf, page 2 -- sleep fraction 20s vs control, zoomed in.
  3. time_window_dsd_rebound.pdf, page 3 -- quiescence fraction 20s vs control.
  4. time_window_dsd_rebound.pdf, page 4 -- quiescence fraction 20s vs control, zoomed in.
  5. time_window_dsd_rebound.pdf, page 5 -- absolute number of stimuli 20s vs control.
  6. time_window_dsd_rebound.pdf, page 6 -- absolute number of stimuli 20s vs control, zoomed in.
  7. time_window_dsd_rebound.pdf, page 9 & 10 -- sleep amount in the 3 (page 15) or 6 hours following SD (ZT 0-[3|6]). Bootstrap error bars.
  8. time_window_dsd_rebound.pdf, page 15 -- observed quiescence - predicted quiescence in 3h (in min). Groups are to be compared to red line, the expectation. Stats are uncorrected Wilcoxon (unpaired) tests. #TODO express in min #TODO add some alpha
  9. time_window_dsd_rebound.pdf, page 16 -- Probability of having a rebound (as defined above) > 0 in 3h (in min). Groups are to be compared to red line (50%), the expectation. Stats are uncorrected Binomial tests.

Sup fig B: Static prolonged SD

• summary : Effect of 9.5 days sleep deprivation 20s intervals.
• experiment -- 20170209_prolonged_sd
• script -- rebound.R
• N > 70 / treatment * sex (see last page)
• figure material:
  1. prolonged_sd_rebound.pdf, page 1 -- sleep fraction 20s vs control.
  2. prolonged_sd_rebound.pdf, page 2 -- sleep fraction 20s vs control, zoomed in.
  3. prolonged_sd_rebound.pdf, page 3 -- quiescence fraction 20s vs control.
  4. prolonged_sd_rebound.pdf, page 4 -- quiescence fraction 20s vs control, zoomed in.
  5. prolonged_sd_rebound.pdf, page 5 -- absolute number of stimuli 20s vs control.
  6. prolonged_sd_rebound.pdf, page 6 -- absolute number of stimuli 20s vs control, zoomed in.
  7. prolonged_sd_rebound.pdf, page 8 -- relative number of stimuli (max = one each 20s) for the 20s interval group. wrapped over circadiand day, from t = days(1).
  8. prolonged_sd_rebound.pdf, page 9 & 10 -- sleep amount in the 3 (page 15) or 6 hours following SD (ZT 0-[3|6]). Bootstrap error bars.
  9. prolonged_sd_rebound.pdf, page 15 -- observed quiescence - predicted quiescence in 3h (in min). Groups are to be compared to red line, the expectation. Stats are uncorrected Wilcoxon (unpaired) tests. #TODO express in min #TODO add some alpha
  10. prolonged_sd_rebound.pdf, page 16 -- Probability of having a rebound (as defined above) > 0 in 3h (in min). Groups are to be compared to red line (50%), the expectation. Stats are uncorrected Binomial tests.
  11. prolonged_sd_surv.pdf, page 1 -- Survival curve (incldes animals that died before rebound)

Sup fig C, effect of changing the tube on behaviour conservation

• summary : Changing tube and shuffeling location => sleep pattern conserved!
• experiment -- 20170109_tube_change_and_var
• script -- script.R
• N > 200 / sex
• figure material:
  1. tube_change_correlations.pdf, page 1 -- sleep fraction after vs before tube shcnage/shuffeling.
  2. tube_change_correlations.pdf, page same as 1, but for quiescence.

Undiscussed figures

Fig X, Sleep deprivation, 12h in L phase.

• summary : Effect of dynamic sleep deprivation from zt 00 to 12.
• experiment -- 20171017_dsd_l_phase
• script -- rebound.R
• N > 60 / treatment * sex (see last page)
• figure material:
  1. l_phase_dsd_rebound.pdf, page 1 -- sleep fraction 20s vs control.
  2. l_phase_dsd_rebound.pdf, page 2 -- sleep fraction 20s vs control, zoomed in.
  3. l_phase_dsd_rebound.pdf, page 3 -- quiescence fraction 20s vs control.
  4. l_phase_dsd_rebound.pdf, page 4 -- quiescence fraction 20s vs control, zoomed in.
  5. l_phase_dsd_rebound.pdf, page 5 -- absolute number of stimuli 20s vs control.
  6. l_phase_dsd_rebound.pdf, page 6 -- absolute number of stimuli 20s vs control, zoomed in.
  7. l_phase_dsd_rebound.pdf, page 9 & 10 -- TODO
  8. l_phase_dsd_rebound.pdf, page 15 -- TODO
  9. l_phase_dsd_rebound.pdf, page 16 -- TODO

Fig Y, Sleep deprivation, 8h in L phase.

• summary : Effect of dynamic sleep deprivation from zt 00 to 8.
• experiment -- 20171017_dsd_zt0_8
• script -- rebound.R
• N > 60 / treatment * sex (see last page)
• figure material:
  1. dsd_zt0_8_rebound.pdf, page 1 -- sleep fraction 20s vs control.
  2. dsd_zt0_8_rebound.pdf, page 2 -- sleep fraction 20s vs control, zoomed in.
  3. dsd_zt0_8_rebound.pdf, page 3 -- quiescence fraction 20s vs control.
  4. dsd_zt0_8_rebound.pdf, page 4 -- quiescence fraction 20s vs control, zoomed in.
  5. dsd_zt0_8_rebound.pdf, page 5 -- absolute number of stimuli 20s vs control.
  6. dsd_zt0_8_rebound.pdf, page 6 -- absolute number of stimuli 20s vs control, zoomed in.
  7. dsd_zt0_8_rebound.pdf, page 9 & 10 -- TODO
  8. dsd_zt0_8_rebound.pdf, page 15 -- TODO
  9. dsd_zt0_8_rebound.pdf, page 16 -- TODO