A scalable, multi-sample Nextflow wrapper for rrwick/Autocycler.
- Singe or Multi-Sample Support: Run a single sample or process dozens of samples simultaneously using a sample sheet.
- Dynamic Assembler Control: Mix, match, or skip specific assemblers via command-line flags (no code editing required).
- HPC-Ready: Built-in profiles for Sun Grid Engine (SGE) cluster execution and scaled-down local debugging.
- Nextflow (
>= 22.10.0) - Execution Engine: Conda/Mamba
- Autocycler Dependencies: Because Autocycler requires specific database downloads (e.g., for Plassembler), it is required to pre-build the developer's Conda environment and point the pipeline to it.
nextflow run main.nf \
--ont_fastq /path/to/reads.fastq.gz \
--sample_id my_isolate_01 \
--genome_size 5.5m \
--out_dir ./resultsnextflow run main.nf
--sample_sheet manifest.txt
--out_dir ./results
If processing multiple samples, provide a space- or tab-separated text file. Note: Lines starting with # are ignored.
| Sample ID | FASTQ Path | Genome Size (Optional) |
|---|---|---|
| Isolate_A | /path/to/barcode01.fastq.gz | 5.5m |
| Isolate_B | /path/to/barcode02.fastq.gz | 3.2m |
If the genome size is omitted in the sample sheet, the pipeline falls back to the --genome_size global parameter.
The pipeline cannot currently estimate the genome size for you, although rrwick has provided steps in autocycler to do so.
- --sample_sheet : Path to the multi-sample manifest file.
- --ont_fastq : Path to a single FASTQ file (if not using a sample sheet).
- --n_subsample : Number of subsamples to create from the input fastq file. (Default: 4).
- --sample_id : Name of the single sample. (Defaults to the base name of the FASTQ file).
- --out_dir : Output directory for all results. (Default: ./autocycler_results).
- --conda_env : Path to the pre-built conda environment. (Default: $HOME/.conda/envs/autocycler_062).
- --assemblers : Comma-separated list of assemblers to run, or all. (Default: all).
- Available: flye, raven, canu, nextdenovo, miniasm, metamdbg, myloasm, necat, plassembler, redbean
- --skip_assemblers : Comma-separated list of assemblers to exclude. (Default: canu).
- --genome_size : Estimated genome size. Must include 'm' or 'k' suffix (e.g., 5.5m).
- --min_read_depth : Minimum read depth required for subsampling. (Default: 25).
- --max_contigs : Maximum number of contigs allowed during compression. (Default: 25).
Profiles manage the execution infrastructure.
- standard (Default): Submits jobs to an SGE HPC cluster. Automates queue routing (short.q, long.q) based on tool time requirements and manages dynamic memory padding (h_vmem).
- local: Executes all jobs on the current node/machine. Caps max resources to prevent system lockups (Max: 16 GB Memory, 8 CPUs).
This pipeline is a Nextflow wrapper designed to automate Autocycler. All core assembly, trimming, clustering, and consensus logic is performed by the Autocycler software.
If you use this pipeline in your research, please cite the original Autocycler manuscript:
Wick RR, Holt KE. Autocycler: generating a robust long-read consensus assembly from multiple diverse assemblies. Bioinformatics. 2024;40(8):btae506. doi: 10.1093/bioinformatics/btae506
- Repository: rrwick/Autocycler
- License: GNU General Public License v3.0
Additionally, please ensure you cite the specific assembler tools (e.g., Flye, Raven, Canu) that you selected to run during your pipeline execution.