Citation
This dataset accompanies the manuscript:
Duncan MJ, Hawkins MR, Bytyqi L, Whitlock HR, Shepard SM, Cox MF, Drinkard EG, Macheda T, Roberts KN, Kohler K, Schmidt MC, Johnson CE, Sunderam S, O'Hara BF, Murphy MP, Bachstetter AD.
Dim light at night impacts circadian rhythms and Alzheimer's disease-like neuroinflammation and neuropathology in humanized APP SAA knock-in mice.
Sleep. 2026 Jun 15;49(6):zsag041. doi: 10.1093/sleep/zsag041. PMID: 41717780.
A preprint version is available at bioRxiv, DOI: 10.1101/2025.10.02.680043.

Author Affiliations

1. Department of Neuroscience, University of Kentucky, Lexington, KY, USA
2. Spinal Cord and Brain Injury Research Center, University of Kentucky, Lexington, KY, USA
3. Georgetown College, Georgetown, KY, USA
4. Department of Molecular & Cellular Biochemistry, University of Kentucky, Lexington, KY, USA
5. Department of Biomedical Engineering, University of Kentucky, Lexington, KY, USA
6. Department of Biology, University of Kentucky, Lexington, KY, USA
7. Sanders-Brown Center on Aging, University of Kentucky, Lexington, KY, USA

Overview
This dataset contains source data corresponding to results presented in the manuscript. The study investigated the effects of dim light at night (dLAN) on circadian rhythms, amyloid-beta (Aβ) pathology, glial reactivity, and neuroinflammation in humanized APP knock-in mice.

Experimental Design
Species: Mus musculus
Strains: hAPPWT knock-in, hAPPSAA knock-in
Sex: Male and female
Age: 12–13 months
Light conditions:
LD = 12:12 light:dark cycle (200 lux light, complete darkness at night)
dLAN = 12:12 cycle with dim light (~5–8 lux) during the dark phase
Duration: 8 weeks exposure
Groups:
hAPPWT-LD, hAPPWT-dLAN, hAPPSAA-LD, hAPPSAA-dLAN

In the data file, lighting condition is recorded in the tx column, where Dark = LD and Dim = dLAN. Genotype is recorded in the geno column, where WT = hAPPWT knock-in and SAA = hAPPSAA knock-in.

Data Files

1. Duncan2025_dLAN_APP-SAA_dataset.csv
   • Subject-level data for all figures in the manuscript
   • Includes circadian measures, Aβ immunohistochemistry, Aβ ELISA, GFAP, CD45, MHCII, and cytokine/chemokine concentrations
   • 39 animals (rows) by 96 variables (columns)
2. Duncan2025_dLAN_APP-SAA_dataset_DataDictionary.csv
   • Detailed data dictionary defining each variable, units, permitted values, and notes
   • One row per variable in the data file; variable names match exactly

Key Variables

• Column names below are as they appear in the data file. Pathology, glial, and inflammatory measures carry a cortex or hippo suffix indicating brain region.
• Identifiers: exp_id (animal ID), tx (lighting condition), sex, geno (genotype), RRID (strain RRID)
• Circadian metrics, each measured at baseline and weeks 2, 4, 6, and 8 (suffixes baseline, wk2, wk4, wk6, wk8): amp* (amplitude), mesor (MESOR), iv (intradaily variability), is (interdaily stability), darkcount and lightcount (activity counts by phase), ldratio (light:dark ratio)
• Amyloid-beta pathology: abeta6e10_area_percent_ (6E10+ percent area), abeta6e10_number_plaques_ (plaque counts by size bin), abeta_fmol_per_mg_pbs_, abeta_fmol_per_mg_tper_, abeta_fmol_per_mg_fa_ (PBS-, detergent-, and formic acid-soluble Aβ, fmol/mg protein)
• Glial reactivity: gfap_area_percent_, cd45_area_percent_, cd45_cluster_count_, mhcii_area_percent_, mhcii_count_
• Inflammatory markers: il1b_, il6_, tnf_, il10_, il17a_, il33_, ccl2_, ccl3_, cxcl1_, cxcl2_, cxcl10_

Reuse Notes

• All values are reported at the individual-animal level
• Sex was included in analysis; no significant sex effects observed
• Data were collected and analyzed blind to group
• Group sizes differ slightly due to animal availability
• In the abeta6e10_number_plaques_ columns, 0 indicates that no plaques were detected. All hAPPWT animals have counts of 0, consistent with the absence of amyloid pathology in this genotype.

Missing Data
Empty cells have been replaced with "NA". NA denotes not available: data were not obtained for that animal and measure. It does not indicate a value of zero, a value below the limit of detection, or a measure that was not applicable to that animal.

Seven of 39 animals have missing values, distributed across both genotypes, both lighting conditions, and both sexes.

For six animals, immunohistochemical measures (6E10, GFAP, CD45, MHCII) are missing because sections were damaged during processing and were excluded from quantification rather than analyzed. Because these markers were quantified from separate stained section series, missingness is specific to the affected series and does not propagate to other markers for the same animal.

For one animal, the hippocampal Aβ ELISA and hippocampal cytokine/chemokine measures are missing because insufficient hippocampal tissue remained for homogenate assays.

Missing values are unrelated to the magnitude of the measurement and were identified before group unblinding. All statistical analyses used the available cases for each measure; no values were imputed. Group sizes therefore differ modestly among measures, as noted under Reuse Notes.

Columns containing NA values: cd45_area_percent_cortex, cd45_area_percent_hippo, cd45_cluster_count_large_cortex, cd45_cluster_count_medium_cortex, cd45_cluster_count_small_cortex, cd45_cluster_count_total_cortex, cd45_cluster_count_large_hippo, cd45_cluster_count_medium_hippo, cd45_cluster_count_small_hippo, cd45_cluster_count_total_hippo, mhcii_area_percent_hippo, mhcii_count_large_cortex, mhcii_count_medium_cortex, mhcii_count_small_cortex, mhcii_count_large_hippo, mhcii_count_medium_hippo, mhcii_count_small_hippo, abeta6e10_area_percent_hippo, abeta6e10_number_plaques_large_hippo, abeta6e10_number_plaques_medium_hippo, abeta6e10_number_plaques_small_hippo, gfap_area_percent_cortex, gfap_area_percent_hippo, abeta_fmol_per_mg_pbs_hippo, abeta_fmol_per_mg_tper_hippo, abeta_fmol_per_mg_fa_hippo, il10_hippo, il1b_hippo, il6_hippo, cxcl1_hippo, tnf_hippo, il33_hippo, cxcl10_hippo, ccl2_hippo, ccl3_hippo, cxcl2_hippo, il17a_hippo.

File Formats
CSV (.csv) for tabular data and variable mapping