Single-nucleus RNA sequencing of male and female brain in a tau P301L transgenic zebrafish: processed and annotated data

Processed and annotated single-nucleus RNA-seq data of whole brains from six-month-old male and female zebrafish: double-transgenic tau P301L fish (Tg(elavl3.2:Gal4-VP16)mde4; Tg(UAS-E1b:DsRed,Hsa.MAPT_P301L)mde3) and wild-type AB controls, two pooled biological replicate libraries per sex–genotype group (three brains per library; eight libraries). Raw reads and per-library Cell Ranger count matrices are available from GEO (GSE310345) and SRA (BioProject PRJNA1365787).

Files: annotated Seurat v5 object (.rds) and AnnData object (.h5ad) of the 118,653 nuclei after quality control and doublet removal (raw and log-normalised counts, SoupX-corrected counts, PCA, UMAP, clusters, cell populations, neurotransmitter-based neuronal layer, atlas label transfer, doublet scores, transgene counts, QC metrics; both objects are provided in parts of 100 MB, to be joined with cat as described in README.md, with md5 checksums of the joined files in MD5SUMS.txt); 10x HDF5 count matrices (raw and SoupX-corrected); per-nucleus metadata (GSE310345_cell_metadata.tsv.gz; barcode = 16-nt barcode-GEO sample number 1–8); SoupX-corrected MEX matrices of all Cell Ranger-called nuclei per library; feature-to-Ensembl mapping; software versions. See README.md.

Processing code: https://doi.org/10.5281/zenodo.23258792 (GitLab: https://gitlab.com/uniluxembourg/lcsb/bds/zebrafish_ad_preprocessing).

Version 2: the transgene columns (tg_Gal4VP16, tg_DsRed, tg_SV40_shared) are based on the sequence of the driver construct: Gal4-VP16 reads are identified by their match to the driver transcript, and reads in the SV40 polyA common to all three transgene transcripts are reported as shared. All other content is unchanged.