CLI Reference

This page summarizes DRAKKAR’s command-line interface. For worked examples and workflow guidance, see User guide, Workflow guide, and Operations and troubleshooting.

Global

$ drakkar <command> [options]

Commands

complete

Run the full pipeline.

preprocessing

Quality filtering and optional host removal.

cataloging

Assembly, binning, and bin metadata.

profiling

Dereplication and abundance profiling.

annotating

Taxonomic and functional annotation.

expressing

Metatranscriptome mapping to annotated genes.

dereplicating

Dereplication only (no read mapping).

inspecting

Microdiversity inspection workflows.

environments

Pre-create Conda environments for modules.

unlock

Unlock a Snakemake working directory.

update

Reinstall DRAKKAR from the Git repository.

transfer

Transfer outputs via SFTP while preserving folder structure.

Selected options

Common options (all workflow modules):

-o/--output: output directory (default: current directory).
-e/--env_path: shared Conda environment directory.
-p/--profile: Snakemake profile (default: slurm).
--overwrite: delete a locked output directory and rerun from scratch.
--skip-benchmark: skip SLURM resource benchmark collection after the run.
--memory-multiplier N: scale per-rule memory requests before the cap.
--time-multiplier N: scale per-rule runtime requests before the cap.

Snakemake override options (all workflow modules):

--snakemake-jobs N: max concurrent SLURM jobs.
--snakemake-cores N: max local CPU cores (local executor only).
--snakemake-executor EXECUTOR: Snakemake executor (e.g. slurm, local).
--snakemake-latency-wait N: seconds to wait for output files.
--snakemake-retries N: retry failed jobs N times.
--snakemake-rerun-incomplete: rerun jobs with incomplete output files.
--snakemake-keep-going: continue after failures.

SLURM override options (all workflow modules):

--slurm-partition NAME: SLURM partition/queue.
--slurm-account NAME: SLURM billing account.
--slurm-constraint EXPR: node constraint expression.
--slurm-nodes N: nodes per SLURM job.
--slurm-nodelist NODES: restrict to specific node(s).
--slurm-extra ARGS: extra sbatch arguments passed verbatim.

Module-specific options are detailed in Workflow guide. Full Snakemake and SLURM management documentation is in Operations and troubleshooting.