Dataset DOI: 10.5061/dryad.d2547d8h0

Description of the data and file structure

This dataset contains confocal microscopy images and amplicon sequencing data characterizing a zebrafish library transgenesis method using delayed site-specific integration. This dataset includes all the raw imaging data associated with Fig. 2, S1, S2, and Table S1, and all the raw Illumina sequencing data associated with Fig. 3, S3, S4, S5, S,6 and Tables S2, S3.

Files and variables

GFP-mScarlet Fluorescent Library Imaging

Format: .nd2 (Nikon raw imaging format, confocal hyperstacks).

Experimental setup: pIGLET embryos were injected wita h 50:50 mix of GFP-CAAX and mScarlet-CAAX plasmids. Larvae imaged at 3 dpf or 5 dpf, live or fixed in 4% PFA, mounted in 1% agarose. Spinning disk confocal microscopy (Yokogawa CSU-W1 on Nikon Ti) with 10×/40× objectives, 488 nm (GFP) and 561 nm (mScarlet) channels, Z-stacks. Used to quantify integration mutual exclusivity (prevalence of neurons with single vs. multiple transgenes per cell). Associated with Fig. 2, S2, Table S1.

File naming:

Files used for mutual-exclusivity quantification (Table S1): [FishID]_FP-library_[Brain_Region]_[Date_acquired].nd2

Files used to produce representative images (Figure 2, S2):FP-library_[Age]_[Region]_[Date].nd2

Barcoded GFP Library Imaging

Format: .nd2 (Nikon raw imaging format, confocal hyperstacks).

Experimental setup: pIGLET embryos were injected with a library of barcoded GFP-CAAX plasmids (each with a unique 15-nt random barcode). Larvae were raised to 5 dpf and imaged live before genome extraction. Confocal imaging (same setup as above) with 488 nm channel to confirm GFP expression before selecting fish for genomic DNA extraction and sequencing. These 12 fish correspond to the samples in sequencing file 6978. Associated with Fig. S1.

File naming: Fish[N]_barcoded-GFP_25_5_26_25.nd2

Sequencing Data Files

Format: .fastq

Experimental setup: 12 larvae injected with barcoded plasmid library at 1-cell stage were raised to 5 dpf, then genomic DNA was extracted from whole bodies. Junction-spanning PCR amplified integrated barcodes and added 5-nt sample barcodes and Illumina overhangs. The source library was directly amplified with primers adding Illumina overhangs. Sequenced on Illumina NextSeq 2000 (paired-end 2×150 bp). Associated with: Fig. 3, S3, S4, S5, S6, Tables S2, S3.

File naming: [SampleID]_001_S[RunNumber]_R[ReadDirection]_001.fastq

• 6978_001_S58_R2_001.fastq - Fish-derived amplicons (12 fish pooled, multiplexed)
• 7462_001_S101_R1_001.fastq - Source library amplicons

Code/software

The complete Python pipeline for processing the sequencing (.fastq) data is available at: https://github.com/shaharbr/library_transgenesis

The pipeline performs demultiplexing of the 12 fish samples (using the 5-nt sample barcodes) from sample 6978, extraction of 15-nt variable barcodes (via flanking anchor sequences) from both samples, error correction (Levenshtein distance ≤1 clustering), quality filtering (≥3 reads/fish, ≥2 reads for source library), and diversity analysis. Reproduces all sequencing figures (Fig. 3D-F, S3-S6) and tables (Tables S2-S3).

The raw imaging files (.nd2) can be opened with ImageJ/Fiji using the Bio-Formats plugin.