Bcl2fastq software v2

Sequencing was performed on an Illumina^® NovaSeq (Illumina, San Diego, CA, USA) with a read length configuration of 150 paired-ends for 120 M paired-ends reads (60 M in each direction) per sample. Run files were demultiplexed using bcl2fastq Software v2.20 (Illumina, San Diego, CA, USA). A quality check was performed on the raw data, removing low quality portions of NGS reads. The trimming step was performed with the following parameters: the minimum length was set to 35 bp and the quality score to 25 using the BBDuk Software. The quality before and after trimming was assessed with the software FASTQC.
HISAT-v2.1.0 was used to map the sequenced reads against an in-house generated from the University of California Santa Cruz (UCSC) reference, which is based on the human reference genome (hg38). After sorting for name and chromosome, followed by indexing with Samtools-v1.9, consistency, and quality of.bam files were checked using Integrative Genomics Viewer-v2.5.3.

Extracted genomic DNA from wild-type or injected embryos were performed by standard protocol. NGS library was constructed using genomic DNA as a template through two rounds of PCR. First-step PCR amplification of 100–270 bp sequences from on/off-target sites was performed using specific primers. For the second-step PCR amplification, PCR products from individual biological samples were amplified using different indexed primers and then pooled into sequencing samples. The samples were subsequently subjected to paired-end read sequencing using the NovaSeq-PE150 strategy at Mingma Technologies Co., Ltd. (Shanghai, China). And the data collection of next-generation sequencing was used by Illumina bcl2fastq Software (v2.20). Finally, the resultant FASTQ files were analysed using CRISPResso2^{25 (link)}. The oligonucleotide sequences used for NGS are listed in Supplementary Data 4.

Sequencing results were demultiplexed and converted to FASTQ format using Illumina bcl2fastq software v2.20 (https://support.illumina.com). Sample demultiplexing, barcode processing, and single-cell 3′ gene counting were carried out using the cell ranger pipeline (https://www.10xgenomics.com/support), and sequencing reads were aligned to Sscrofa11.1 reference genome to obtain the raw digital gene expression matrix with the unique molecular identifier (UMI) counts per gene per cell. The cell ranger outputs were then loaded into Seurat v4.0.3 [155 (link)], and the cell candidates were removed if they expressed fewer than 500 unique genes, more than 50,000 UMI counts, or greater than 25% mitochondrial reads. To visualize the data, we further reduced the dimensionality of all high-quality cells using Seurat, and applied the t-SNE algorithm to project the cells into 2D space. In detail, the “normalization” function in the Seurat package was used to calculate the expression levels of genes; the “FindClusters” function with default parameters was performed to perform cell clustering; the “FindAllMarkers” function was used to determine the DEGs or marker genes with |Log(Fold ChangE)|> 0.25 and P value < 0.05. GO and KEGG annotation of DEGs were performed by the “enrichGO” and “enrichKEGG” functions in the clusterProfiler package v4.0.5 [156 (link)].