Table 2.

Power analysis for the sample size of 4 drivers used in Figure 2^a

Total sleep	Experiment vs GAL4 Control				Experiment vs UAS Control
	30°C		21°C		30°C		21°C
Drivers	LP	DP	LP	DP	LP	DP	LP	DP
c232	1	0.167	0.377	0.639	1	0.563	0.533	0.907
Feb170	1	1	0.946	0.209	1	1	0.996	0.19
R48B10	0.151	0.53	0.934	0.077	1	0.998	0.975	0.999
R53F11	1	0.052	0.999	0.858	0.108	0.996	0.85	0.928

No. of episodes	Experiment vs GAL4 Control				Experiment vs UAS Control
	30°C		21°C		30°C		21°C
Drivers	LP	DP	LP	DP	LP	DP	LP	DP
c232	0.057	0.09	0.497	0.268	0.102	0.347	0.561	0.186
Feb170	0.929	0.99	0.234	0.083	0.819	0.06	0.989	0.86
R48B10	1	0.977	0.985	0.286	1	0.948	0.521	0.287
R53F11	0.816	1	0.948	0.149	0.705	0.992	0.323	0.078

Maximum episode length	Experiment vs GAL4 Control				Experiment vs UAS Control
	30°C		21°C		30°C		21°C
Drivers	LP	DP	LP	DP	LP	DP	LP	DP
c232	0.999	0.182	0.075	0.05	0.975	0.46	0.159	0.057
Feb170	1	0.795	0.053	0.05	0.993	0.859	0.673	0.121
R48B10	0.814	0.051	0.244	0.061	1	0.21	0.502	0.086
R53F11	0.815	0.188	0.275	0.219	0.413	0.091	0.19	0.054

^aFour GAL4 drivers were included in the analysis: c232, Feb170, R48B10, and R53F11. The experimental group was compared with either GAL4 control group or UAS-dTrpA1 control group for both activation day and recovery day. Total sleep, number of episodes, and maximum episode length for LP and DP were analyzed separately. Power analysis was conducted using the script of sampsizepwr in MATLAB.