scHiCAR

Pipelines for scHiCAR data processing

Tools requirement

• Python: python3.7 or later
• snakemake: pip install snakemake==5.13.0
• cutadapt: pip install cutadapt==3.3
• pairtools: pip install pairtools==1.1.0
• snaptools: pip install snaptools==1.4.8
• STAR: v2.7.5c
• BWA: v0.7.17
• SAMTOOLS: v1.12

1. Preprocess raw FASTQ files of the RNA library with Snakemake and align sequences to the genome (code)

a. Snakemake procedures:

• Trim specific sequences at the 5′ end of Read 1.
• Extract RNA barcodes from Read 1 and append them to its 5′ end.
• Remove the template-switching oligo (TSO) from the 5′ end of Read 2.
• Remove poly(A) tails and adaptor sequences from the 3′ end of Read 2.
• Split FASTQ files into two subsets based on primer strategy: oligo-dT vs. random hexamer.
• Extract and count all barcodes from the dataset.
• Correct barcodes that have only one mismatch relative to the whitelist.
• Compress and merge filtered FASTQ files from the two primer strategies.

The resulting files (03_corrected_fq/*_all_L001_R*_001.fastq.gz) are ready for alignment using STAR.

b. Generate filtered gene expression matrices (barcodes.tsv, features.tsv, and matrix.mtx) with STAR.

2. Preprocess raw FASTQ files of the DNA library with Snakemake (code)

Snakemake procedures:

• Trim specific sequences at the 5′ end of both Read 1 and Read 2.
• Extract barcodes from the read sequences and append them to the read names following the @ symbol.
• Generate a list of all extracted barcodes and count their occurrences.
• Compare extracted barcodes against a provided whitelist.
• Correct barcodes that have only one mismatch relative to the whitelist.
• Compress the filtered FASTQ files.
• Remove ME (mosaic end) sequences from the reads.

The resulting files (05_cutME_fq/*_cutME_L001_R*_001.fastq.gz) are ready for generating ATAC fragment files and chromatin contact pair files.

3. Generate ATAC fragment files with Snakemake (*.tsv.gz)(code)

Snakemake procedures:

• Align R1 reads to the reference genome using Snaptools with BWA, and sort BAM files by read name.
• Convert BAM files to fragment-level BED format.
• Extract high-quality cell barcodes based on the knee point of the barcode rank curve.
• Filter fragments using the high-quality barcode list.

The resulting files (03/filtered/*.filtered.tsv.gz) can be used in standard scATAC-seq downstream analysis.

4. Generate chromatin contact pair files with Snakemake (*.dedup.pairs.gz)(code)

Snakemake procedures:

• Align reads using BWA-MEM with -SP mode and convert SAM output to BAM format.
• Parse mapped reads, select valid contact pairs, and retain 28-bp cell barcodes.
• Flip contact pairs to generate an upper-triangular matrix.
• Sort contact pairs by cell barcode and genomic position.
• Remove PCR and optical duplicates.
• Summarize read pair statistics.
• Filter contact pairs based on the knee point of the barcode rank curve.

The resulting files (05_filtered/*.dedup.filtered.pairs.gz) can be used in downsteam pseudo-bulk or single-cell contact analysis.

5. Downsteam pseudo-bulk / single-cell analysis (code)

Citation

Wei, X., Xu, Y., Yang, D. et al. Trimodal single-cell profiling of transcriptome, epigenome and 3D genome in complex tissues with scHiCAR. Nat Biotechnol (2026). https://doi.org/10.1038/s41587-026-03013-7