Introduction Usage Parameters Output

Usage

Input

The pipeline requires a phyloseq RDS object (.RDS) containing:

  • An OTU/ASV table (taxa as rows, samples as columns)
  • A sample metadata data frame with at least a condition/group column
  • A taxonomy table
--input /path/to/phyloseq.RDS

Optional: pre-extracted control phyloseq (Path B only)

If you have already subset the control/healthy samples into a separate phyloseq object, provide it with --input_control. Otherwise, the pipeline extracts them automatically using --condition and --base_level.

--input_control /path/to/control_phyloseq.RDS

Execution paths

Path A: k-intersection consensus (--simulate false)

Runs DA tools on the real dataset and computes a k-intersection consensus.

nextflow run nf-core/daamicrobiome \
  -profile docker \
  --input phyloseq.RDS \
  --simulate false \
  --condition "study_condition" \
  --base_level "healthy" \
  --outdir results/

Path B: simulation-trained weighted consensus (--simulate true)

Default mode. Extracts control samples, simulates ground-truth datasets via MIDASim, scores DA tools, and applies the optimal threshold to real data.

nextflow run nf-core/daamicrobiome \
  -profile docker \
  --input phyloseq.RDS \
  --simulate true \
  --condition "study_condition" \
  --base_level "healthy" \
  --outdir results/

Parameters

Core parameters

Parameter Default Description
--input (required) Path to the full phyloseq RDS object
--input_control null Path to a control-only phyloseq RDS (optional, Path B)
--outdir results/ Output directory
--simulate true false = Path A (k-intersection), true = Path B (weighted consensus)

Metadata parameters

Parameter Default Description
--condition study_condition Column in sample_data identifying the group variable
--base_level healthy Reference/control level in the condition column
--confounder null Optional confounder column(s) for real data DA tools
--bc_confounder null Batch-correction confounder (MaAsLin2 only)

DA tool toggles

All tools are enabled by default except metagenomeSeq.

Parameter Default
--tool_adapt true
--tool_corncob true
--tool_linda true
--tool_locom true
--tool_maaslin2 true
--tool_metagenomeseq false

Simulation parameters (Path B only)

Parameter Default Description
--sim_n_ctrl null (auto) Number of control samples per simulation
--sim_n_case null (auto) Number of case samples per simulation
--sim_p_da [0.05, 0.10] Proportion of DA taxa
--sim_abs_logfc [0.5, 1.0, 2.0] Absolute log-fold change values
--sim_n_rep 3 Number of replicates per scenario
--sim_design ["balanced"] Simulation design
--sim_seed0 123 Random seed for reproducibility

Scoring parameters (Path B only)

Parameter Default Description
--eval_alpha 0.05 Significance threshold
--eval_fdr_target 0.05 Target FDR for threshold optimisation
--eval_fdr_reference 0.10 Reference FDR level
--eval_retention_at_ref 0.25 Minimum retention at reference FDR

Path A parameters

Parameter Default Description
--consensus_alpha 0.05 Significance threshold for k-intersection
--consensus_lfc_min 0 Minimum absolute log-fold change filter

Resource limits

Parameter Default Description
--max_cpus 16 Maximum CPUs per process
--max_memory 128.GB Maximum memory per process
--max_time 240.h Maximum wall time per process

Profiles

Profile Description
docker Run with Docker containers
hpc Run with Singularity on a SLURM cluster
On this page