Description of the data and file structure

Data are available for reuse. The authors have no conflicts to disclose.

Zebrafish sleep analysis output metrics.

Data were collected using ViewPoint Software and first analyzed in MATLAB. Statistical comparisons were performed in SAS.

Eleven sleep and activity parameters were analyzed for each crispant (gene manipulation group) vs. negative control-injected siblings.

All behavioral measures are provided in the file "CRISPR_Allgenes_OutliersMarked." This file contains the averages across the two days in which sleep is captured (days 6 and 7 post-fertilization). All fish from each clutch are included, and fish that were considered extreme outliers by visual inspection are identified. These were likely representative of fish that died or became unhealthy during the assay,y or had debris or a bubble obscuring the camera. 

All statistical comparisons were performed within each gene-control pair. No comparisons were made between different genes. 

Primary comparisons between crispant and negative control-injected siblings were performed using both parametric t-test and non-parametric Wilcoxon rank-sum tests to mitigate any potential impact of non-normality of the endpoints. A Hochberg step-up procedure[1],[2] was applied to the analysis of each gene independently to maintain gene-specific type I error at the desired level of 0.05 across the 11 tested hypotheses. Notably, this multiple testing correction procedure defines a threshold for significance based on p-value ranks. The behavioral measures deemed significant by this threshold are bolded in the file "ZebrafishSleepData_Zimmerman2025_DRYAD."

[1] Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. *Biometrika. *1988:75(4):800-802.

[2] Huang Y and Hsu J.C. Hochberg’s step-up method: cutting corners off Holm’s step-down method. Biometrika. 2007:94(4):965-975

Output:

• n_Control/ n_[Gene name] = sample size
• musd = mean (standard deviation)
• muci =  mean (95% confidence interval of the mean)
• medrng = median (range)
• diffci95 = mean difference (95% confidence interval)
• smd = standardized mean difference (effect size)
• pc_ttest = p-value corrected t-test (assumes normality)
• hochsig_ttest = Did the corrected p-value reach the significance threshold for the Hochberg step-up procedure following multiple t-test correction? 
• pc_rnksum = p-value corrected following rank sum test (does not assume normality) 
• hochsig_rnksum = Did corrected p-value reach significance threshold for Hochberg step-up procedure following rank sum test [this option was used to determine significance, as many activity traits are non-normally distributed]
• Pooled SD = pooled standard deviation - used to calculate SMD 
• SE of SMD = standard error of SMD - used to calculate confidence intervals of SMD
• LI 95 CI = lower interval for the 95% Confidence Interval of the SMD
• UI 95 CI = upper interval for the 95% Confidence Interval of the SMD

All behavioral traces for days 6 and 7 post fertilization,n as well as individual data points, are plotted in the file "Zimmermanetal-SupplementaryBehaviorData."

Data are shown for each clutch separately.

Sanger Sequencing files are provided for each gene target as a sequencing file (.seq) and a Sanger trace (.ab1). These files were used to generate Supplementary Figure 2. Each file represents the guideRNA target amplicon generated using target-specific primers for a single crispant fish, showing the on-target mutation generated by the guideRNA. File names indicate gene name [meis1a, meis1b, cbx1a, cbx1b, gnb3a, gnb3b, skiv2l, tcf12, arfgap2] or negative control sibling [negmeis1b = negative control sibling of meis1b crispant]. Forward or Reverse strand is indicated by (F) or (R). The remaining letters are lane/sample identifiers. (e.g. meis1bF1_A09_079 = meis1b crispant, Forward strand). Corresponding .seq and .ab1 files are labeled with the same name. Sample QC was provided by the University of Pennsylvania Sanger Sequencing Core. No post-processing was applied.