Skip to content

Parameter Reference

Every pipeline parameter, grouped by pipeline stage. nextflow_schema.json at the repo root is the source of truth for types, defaults, and validation.

For a walkthrough of the parameters needed for a first run, see Running the CLI.

-- vs -

Pipeline parameters (listed below) are passed with a double dash, e.g. --aligner MAFFT. Nextflow's own built-in options use a single dash, e.g. -profile, -output-dir, -c.

Core parameters

Required for every run.

Parameter Type Default Description
run_name string (required) Identifier for this run. Used in logging, reports, and the default output path.
samplesheet file path (required) CSV of samples and metadata. See Samplesheet Reference.
sample_base_dir directory path (required) Base directory for resolving samplesheet file paths.
reference_file file path (required) Reference sequence containing the gene/CDS of interest. Expected format depends on trim_method.
region_of_interest string (required) CDS or protein to extract. Required even without AGA preprocessing; no fixed naming convention.
-output-dir directory path ./results/${run_name} Nextflow built-in, not a pipeline parameter. Where output files are saved.

Preprocessing parameters

Stage 1 (Trim) — see Pipeline Overview.

Parameter Type Default Description
trim_method enum: MINIMAP2, AGA AGA Trimming tool. MINIMAP2 is faster; AGA handles complex overlapping reading frames.
skip_trim boolean false Skip trimming — input is already trimmed to the region of interest.
region_shorthand string (none) Short name for the region of interest.
aga_args string --local Extra arguments for AGA.
minimap_trim_from integer (none) Start coordinate for MINIMAP2 trimming.
minimap_trim_to integer (none) End coordinate for MINIMAP2 trimming.

Filtering parameters

Stage 2 (Functional Filtering) — see Pipeline Overview.

Parameter Type Default Description
functional_filter_method enum: ELLPACA, LENGTH_BASED_FILTERING ELLPACA ELLPACA filters on stop codons (below); LENGTH_BASED_FILTERING filters on length (next section).
skip_functional_filter boolean false Skip functional filtering — all sequences proceed to alignment.
ff_max_stop_pct integer 100 Latest point along the read (as a percentage) a stop codon can occur and still count as functional.
ff_include_no_stop_codons boolean true Whether sequences with no stop codons are functional.
ff_include_frameshifts boolean false Whether sequences with a potential frameshift are functional.
ff_acceptable_pct_loss number (0–1) 0.2 Proportion of median protein length allowed to be lost before a sequence is non-functional.
ff_expected_length number (median of input) Expected protein length. Unset uses the median of the input.

Length-based filtering

Used when functional_filter_method is LENGTH_BASED_FILTERING.

Parameter Type Default Description
lbf_threshold_mode enum: FIXED, MEDIAN, MEAN MEDIAN How the minimum-length threshold ("center") is determined.
lbf_min_length integer (none) Minimum length to keep. Required, and only used, in FIXED mode.
lbf_tolerance string (e.g. 20, 20%) (none — strict minimum) Sets both lbf_min_tolerance and lbf_max_tolerance to the same value. Not combinable with either.
lbf_min_tolerance string (e.g. 20, 20%) (none) How much shorter than center a sequence may be. Not combinable with lbf_tolerance.
lbf_max_tolerance string (e.g. 20, 20%) (none — no upper bound) How much longer than center a sequence may be. Not combinable with lbf_tolerance.

K-mer filtering

Independent filtering pass matching sequence ends against expected k-mers.

Parameter Type Default Description
use_kmer_filtering boolean false Turns k-mer filtering on.
match_kmers_start string (none) K-mer(s) to match at the sequence start. Comma-separated.
match_kmers_end string (none) K-mer(s) to match at the sequence end. Comma-separated.

Alignment parameters

Stage 3 (Alignment) — see Aligner Reference for choosing an aligner.

Parameter Type Default Description
aligner enum (see Aligner Reference) MAFFT Aligner used for the main alignment.

Pass-through arguments, one per aligner:

Parameter Default Applies to
mafft_args --localpair --maxiterate 1000 MAFFT
mafft_fast_align_args --retree 1 --maxiterate 0 MAFFT, used internally by AGA's preprocessing
tcoffee_args (none) T-Coffee
tcoffee_regressive_args -nseq 100 -tree mbed -method clustalo_msa T-Coffee Regressive
prank_args +F PRANK
pagan_args (none) PAGAN
clustalo_args --iterations 2 Clustal Omega
clustalw_args -CLUSTERING=NJ -ITERATION=NONE -NUMITER=3 -MATRIX=BLOSUM ClustalW
virulign_args --maxFrameShifts 3 --exportReferenceSequence no VIRULIGN
macse_args (none) MACSE

MUSCLE, MUSCLE-FAST, and PROBCONS have no pass-through arguments parameter.

Panel alignment

Parameter Type Default Description
panel_alignment file path (none) Pre-aligned sequences used as a scaffold, when aligner is MAFFT-SEED.

Reference sequence insertion

Inserts a reference sequence directly into the alignment, instead of mapping it on afterward.

Parameter Type Default Description
add_reference_to_sequences enum: BEFORE, AFTER, null null Whether, and where, a reference is added to the sequences being aligned. null disables this.
reference_to_add file path (none) Reference sequence to insert. Can differ from reference_file, which is used for trimming/mapping.

Phylogeny parameters

Infers a phylogenetic tree from the computed alignment(s).

Parameter Type Default Description
build_phylogeny boolean false Infer a phylogenetic tree from the computed alignment(s).
phylogeny_alignment_type enum: NT, AA, BOTH NT Which computed alignment(s) feed tree inference.
phylogeny_baseline_method enum: REFERENCE, CONSENSUS, MINDIST REFERENCE How the baseline sequence for tree rooting is chosen. Also intended for a per-site variation heatmap, not yet implemented.
iqtree_args string (none) Extra arguments for IQ-TREE, when build_phylogeny is enabled.

Deprecated

multi_timepoint_alignment is deprecated. Don't use it in new runs.

Execution / resource parameters

Parameter Type Default Description
-profile string (none) Nextflow config profile(s). Container runtimes: docker, singularity, apptainer. Others: test (caps resources for the test dataset), debug (verbose logging, no cleanup), slurm. Site-specific: ilifu, hex, zoidberg — see the Developer Guide to add your own.
max_memory string 128.GB Memory cap per process.
max_cpus integer 16 CPU cap per process.
max_time string 240.h Time cap per process.
slurm_queue string Main SLURM queue, when using the slurm profile.