Worked analyses¶
These examples show how Aviary stages connect. Replace the illustrative paths with real input files; no example dataset is bundled with the repository.
Hybrid metagenome from reads to MAGs¶
Input¶
Preview¶
aviary complete \
-1 reads/sample_R1.fastq.gz \
-2 reads/sample_R2.fastq.gz \
--longreads reads/sample_ont.fastq.gz \
--long-read-type ont \
--output hybrid_complete \
--max-threads 16 \
--n-cores 32 \
--dry-run
Run¶
Remove --dry-run after checking the planned targets and database paths:
aviary complete \
-1 reads/sample_R1.fastq.gz \
-2 reads/sample_R2.fastq.gz \
--longreads reads/sample_ont.fastq.gz \
--long-read-type ont \
--output hybrid_complete \
--max-threads 16 \
--n-cores 32 \
--max-memory 250
Interpret¶
Inspect assembly/final_contigs.fasta for the final assembly,
bins/bin_info.tsv for the MAG summary and bins/final_bins/ for genome FASTA
files. Check logs/ for rule execution details and benchmarks/ before
adjusting scheduler requests.
Separate assembly and recovery¶
Splitting stages is useful when you want to inspect the assembly before committing to genome recovery.
aviary assemble \
-1 reads/sample_R1.fastq.gz \
-2 reads/sample_R2.fastq.gz \
--output 01_assembly \
--max-threads 16 \
--n-cores 16
aviary recover \
--assembly 01_assembly/assembly/final_contigs.fasta \
-1 reads/sample_R1.fastq.gz \
-2 reads/sample_R2.fastq.gz \
--output 02_recovery \
--max-threads 16 \
--n-cores 32
The reads are supplied again because coverage across the assembly contributes to genome recovery and abundance estimation.
Annotate an existing genome collection¶
aviary annotate \
--genome-fasta-directory 02_recovery/bins/final_bins \
--fasta-extension fna \
--output 03_annotation \
--max-threads 16 \
--n-cores 16
Keep the annotation run with the reference database releases used. See the annotation guide for interpretation and the CLI reference for advanced controls.