CORRAL

A Nextflow pipeline that aligns sequencing reads against a marker-gene reference database and summarizes taxon-level abundance from the alignments.

Overview

This pipeline aligns short reads to a marker-gene reference database with bowtie2, then summarizes the resulting alignments into taxon abundance tables using marker_alignments. It is used within VEuPathDB's metagenomic/eukaryotic-community profiling workflows to estimate the taxonomic composition of a sample from marker-gene read alignments. Reads can be pulled from SRA by run accession or read directly from local FASTQ files; for each sample, reads are aligned to a reference database, per-alignment statistics are computed, alignments are summarized into per-sample taxon tables, and the per-sample tables are finally combined into one summary matrix.

Requirements

• Nextflow (DSL2)
• Docker (enabled by default via nextflow.config)

The pipeline runs on the veupathdb/corral:latest container image, which bundles bowtie2, samtools, the SRA toolkit (fasterq-dump), and the marker_alignments package.

Usage

nextflow run VEuPathDB/CORRAL \
  -r main \
  -entry sra \
  --inputPath /path/to/runAccessions.tsv \
  --refdb /path/to/database/refdb \
  --markerToTaxonPath /path/to/marker-to-taxon-id.tsv \
  --libraryLayout paired \
  --resultDir /path/to/output \
  -resume

The pipeline has two named workflow entry points:

• sra — reads a TSV of run accessions from params.inputPath (a run_accession column), downloads each run's reads from SRA with fasterq-dump, then aligns, scores, and summarizes them.
• local — reads paired FASTQ files directly from the directory given in params.inputPath (matched as *_{1,2}.fastq), then aligns, scores, and summarizes them without a download step.

Both entry points converge on the same alignment/summarization logic: bowtie2 alignment, per-sample alignment statistics, per-sample taxon summarization, and a final combined summary table.

Key Parameters

Parameter	Description
params.downloadMethod	sra or local; selects whether reads are downloaded from SRA or read from local files (used only when running without an explicit -entry).
params.inputPath	For sra: path to a TSV listing run_accession values. For local: path to a directory of paired FASTQ files.
params.libraryLayout	single or paired; selects which bowtie2 alignment template is used.
params.refdb	Path prefix passed to bowtie2 -x for the marker-gene reference database.
params.bowtie2Command	The bowtie2 command line (including flags) used for alignment.
params.alignmentStatsCommand	Command used to compute alignment statistics (e.g. samtools stats).
params.summarizeAlignmentsCommand	The marker_alignments command line (including filter thresholds) used to summarize alignments into taxon calls.
params.markerToTaxonPath	Path to the marker-to-taxon-ID mapping TSV used by marker_alignments.
params.summaryColumn	Value column used when building the final summary matrix (e.g. cpm).
params.summaryFormat	Output format for the final summary table (e.g. matrix).
params.resultDir	Directory where alignment stats, per-sample taxon summaries, and the final combined table are published.

Output

Under params.resultDir: an alignmentStats/ directory with per-sample alignment statistics, a summarizedAlignments/ directory with per-sample <sample>.taxa.tsv taxon summaries, and a combined <summaryColumn>.<summaryFormat>.tsv file with taxon abundance across all samples.