Quickstart

This example runs the complete short-read metagenome workflow. It assumes Aviary is installed and the required databases are configured.

1. Check the installation

aviary --version
aviary complete --full-help

2. Prepare paired reads

Use one forward and one reverse FASTQ file for the same sample:

reads/
├── sample_R1.fastq.gz
└── sample_R2.fastq.gz

Compressed FASTQ input is accepted. Forward and reverse files must be supplied in matching order when more than one pair is given.

3. Preview the workflow

aviary complete \
  -1 reads/sample_R1.fastq.gz \
  -2 reads/sample_R2.fastq.gz \
  --output sample_aviary \
  --max-threads 8 \
  --n-cores 8 \
  --dry-run

The dry run resolves the workflow without executing analysis tools. Remove --dry-run after checking the planned jobs and configured database paths.

4. Run

aviary complete \
  -1 reads/sample_R1.fastq.gz \
  -2 reads/sample_R2.fastq.gz \
  --output sample_aviary \
  --max-threads 8 \
  --n-cores 8

--max-threads limits an individual tool. --n-cores is the total CPU capacity available to Snakemake, which may schedule several compatible jobs at once.

5. Inspect the results

Start with:

sample_aviary/
├── assembly/final_contigs.fasta
├── bins/bin_info.tsv
├── bins/final_bins/
├── benchmarks/
└── logs/

Read bin_info.tsv and the rest of the output layout before using recovered genomes downstream. If a rule fails, find its log in sample_aviary/logs/ and see troubleshooting.

Next step

The first complete analysis explains how the stages behave and which controls are most important.