seq-wrangler

🧬 Seq Wrangler

ClawBio skill — FASTQ QC, alignment, and BAM processing pipeline for NGS data.

Part of the ClawBio bioinformatics AI agent skill library.

Quick Start

# Demo (no external tools needed)
python skills/seq-wrangler/seq_wrangler.py --demo --output /tmp/demo

# Single sample paired-end
python skills/seq-wrangler/seq_wrangler.py \
    --r1 sample_R1.fastq.gz \
    --r2 sample_R2.fastq.gz \
    --index ref/hg38 \
    --aligner bowtie2 \
    --output results/

# Batch mode via samplesheet
python skills/seq-wrangler/seq_wrangler.py \
    --samplesheet samples.csv \
    --index ref/hg38 \
    --output results/

What it produces

output/
├── report.md # Full alignment and QC report
├── summary.json # Per-sample stats as JSON
├── bam/
│ └── sample_sorted.bam # Final sorted, markdup BAM + .bai index
├── alignment/
│ └── sample.sam # Intermediate SAM (only with --keep-sam)
└── reproducibility/
├── commands.sh # Reproduce this exact run
├── environment.yml # Conda environment
├── checksums.sha256 # SHA-256 of all input files
└── run_metadata.json # Full run parameters

Options

Flag	Description	Default
--r1	FASTQ R1 or single-end FASTQ	—
--r2	FASTQ R2 for paired-end	—
--samplesheet	CSV with columns sample,fastq1,fastq2	—
--index	Aligner index prefix	required
--aligner	bwa, bowtie2, or minimap2	bwa
--genome-build	GRCh38 or GRCh37	GRCh38
--output	Output directory	required
--threads	Number of threads	auto
--mapq	MAPQ filter threshold	20
--trim	Run fastp trimming before alignment	off
--remove-duplicates	Remove duplicates with samtools markdup -r	off
--keep-sam	Keep intermediate SAM files	off
--run-fastqc	Run FastQC if available	off
--run-multiqc	Run MultiQC aggregation if available	off
--demo	Run with synthetic data, no tools needed	off

Pipeline steps

1. FastQC — per-sample quality assessment (optional)
2. fastp — adapter trimming and QC (optional, --trim)
3. Alignment — BWA MEM / Bowtie2 / Minimap2
4. samtools view — filter by MAPQ threshold
5. samtools sort -n — sort by read name
6. samtools fixmate — fix mate-pair information
7. samtools sort — coordinate sort
8. samtools markdup — mark (or remove) PCR duplicates
9. samtools index — index final BAM
10. samtools flagstat / coverage / stats — alignment statistics
11. MultiQC — aggregated QC report (optional)
12. Report — Markdown report + reproducibility bundle

Dependencies

Install via conda (recommended):

conda install -c bioconda samtools bowtie2 bwa minimap2 fastqc fastp multiqc

Or use the included environment file:

conda env create -f skills/seq-wrangler/reproducibility/environment.yml
conda activate clawbio-seq-wrangler

Run Tests

python -m pytest skills/seq-wrangler/tests/ -v

Author

Contributed by Daniel Garbozo — resolves Issue #10.