csplice 0.1.4

Genomic data processing toolkit

csplice

Genomic data processing toolkit for converting GTF to BED format and analyzing BAM gene overlaps.

Installation

# Install from PyPI
pip install csplice

# Or install from source
pip install .

Usage

gtf2bed command

Convert GTF to BED files:

csplice gtf2bed -g input.gtf -o output_dir

This will generate 4 BED files in the output directory:

• gene.bed: Standard 6-column BED format (chrom, start, end, gene_id, gene_name, strand)
• transcript.bed: Standard 6-column BED format (chrom, start, end, transcript_id, gene_id, strand)
• exon.bed: Standard 6-column BED format (chrom, start, end, transcript_id, gene_id, strand)
• intron.bed: Standard 6-column BED format (chrom, start, end, transcript_id, gene_id, strand)

bam2gene command

Analyze BAM file gene overlaps:

csplice bam2gene -b input.bam -g genes.bed -o output_dir

This will generate 2 files in the output directory:

• splice.txt: TSV format with columns (barcode, umi, gene_id, gene_name), reads with out intron
• unsplice.txt: TSV format with columns (barcode, umi, gene_id, gene_name), reads with intron

Options

gtf2bed

Options:
  -g, --gtf TEXT         Input GTF file path  [required]
  -o, --outdir TEXT      Output directory  [required]
  -i, --gene_id TEXT     Gene ID key in attributes  [default: gene_id]
  -n, --gene_name TEXT   Gene name key in attributes  [default: gene_name]
  -t, --transcript_id TEXT  Transcript ID key in attributes  [default: transcript_id]
  --help                 Show this message and exit.

bam2gene

Options:
  -b, --bam TEXT         Input BAM file path  [required]
  -o, --outdir TEXT      Output directory  [required]
  -g, --genebed TEXT     Gene BED file path  [required]
  -i, --introned TEXT    Intron BED file path  [required]
  -c, --cb TEXT          Cell barcode tag (default: CB)
  -u, --ub TEXT          UMI tag (default: UB)
  --help                 Show this message and exit.

Development

# Install dev dependencies
poetry install --with dev

# Run tests
poetry run pytest