Configuration

Configuration

All parameters are set in config/default.yaml. It is recommended to copy the file and pass it with --configfile rather than editing it directly.

---

Pipeline configuration

Option	Default	Choices	Description
vcf2phylip	False	True / False	Use multi-sample VCF instead of BUSCO-based pipeline
quastcore	True	True / False	Compute assembly statistics
alignment	"mafft"	mafft, muscle, prank	Multiple alignment tool
filtration	"clipkit"	clipkit, trimal, gblocks	Alignment trimming tool
iqtree	True	True / False	Run IQ-TREE
astral	True	True / False	Run ASTRAL-IV (multispecies coalescent)
rapidnj	True	True / False	Run RapidNJ
phylip	True	True / False	Run PHYLIP
raxml	True	True / False	Run RAxML-NG
mrbayes	False	True / False	Run MrBayes (recommended to run GPU-compiled version separately)
draw_phylotrees	True	True / False	Visualize output trees

---

Tool parameters

BUSCO

Parameter	Default	Description
busco_dataset_path	"path/to/busco_datasets/ortho_odb12/"	Path to pre-downloaded OrthoDB dataset
busco_options	"--offline"	Extra BUSCO flags; use "--offline" to run without internet
busco_mode	"genome"	BUSCO mode
busco_blacklist	"input/BUSCO.blacklist"	File with BUSCO IDs to exclude (optional)
busco_histogram_colors	"#23b4e8,#008dbf,#fbbc04,#ea4335"	Bar colors for S, D, F, M in the BUSCO histogram

Reconstruction (VCF SNPs and/or consensus FASTAs)

Reconstructs per-sample BUSCO sequences from a reference's BUSCO output. Input goes under input/reconstruct/<subdir>/ — one reference FASTA at the subdir top level plus optional vcf/ (per-sample .vcf.gz → .AltRef) and fasta/ (consensus genomes → .Consensus) directories. See Usage#reconstructed-samples-against-a-reference-genome.

Parameter	Default	Description
apply_vcf_iupac	False	(VCF source only) Encode heterozygous SNPs as IUPAC ambiguity codes (equivalent to GATK --use-iupac-sample); if False, ALT allele is used for het/hom-alt calls
apply_vcf_random_het	False	(VCF source only) Resolve each genotype by drawing one called allele at random, so heterozygous sites collapse to a single REF or ALT base. Mutually exclusive with apply_vcf_iupac
apply_vcf_seed	0	Random seed used by apply_vcf_random_het so the draws are reproducible across reruns
reconstruct_refs_as_species	False	Include each reference genome itself as a sample in the output phylogeny
altref_gapaware_insertion	False	Insert reconstructed sequences into alignments using gap positions of the corresponding reference instead of re-aligning; recommended when working with many reconstructed samples; see Advanced-Usage#gap-aware-insertion-of-reconstructed-sequences

Alignment

Parameter	Default	Description
mafft_params	"--reorder --auto"	Flags passed to MAFFT
muscle_params	""	Flags passed to MUSCLE
prank_params	"-codon"	Flags passed to PRANK
prank_time	"100h"	PRANK is killed 15 min before this limit; genes that time out are discarded

Filtration

Parameter	Default	Description
clipkit_params	"--mode smart-gap"	Flags passed to ClipKIT; add --codon for codon-aware trimming
trimal_params	"-automated1"	Flags passed to TrimAl
gblocks_params	"-t=Codons"	Flags passed to GBlocks

Phylogenetic inference

Parameter	Default	Description
iqtree_params	"-keep-ident -m TESTNEW -bb 1000"	IQ-TREE flags for the concatenated alignment; add -o 'OUTGROUP' to set an outgroup
iqtree_per_fna_params	"-keep-ident -m TESTNEW -bb 1000"	IQ-TREE flags for per-gene trees (used as input to ASTRAL)
nodes_filtration_by_support	70	Minimum bootstrap support for collapsing nodes before ASTRAL
astral_params	"--support 2"	ASTRAL-IV flags; add --root 'OUTGROUP' to set an outgroup
raxml_params	"--model GTR+G --bs-trees 100"	RAxML-NG flags
rapidnj_params	"-b 1000"	RapidNJ flags
phylip_dnadist_params	"D\n"	"D\n" for Kimura 2-parameter model; "" for F84 (default)
phylip_neighbor_params	""	"N\n" for UPGMA; "" for NJ (default)
mrbayes_params	""	Extra MrBayes flags
mrbayes_block	"resources/mrbayes.block"	Path to MrBayes block configuration file
mrbayes_path	"path/to/mrbayes-3.2.7/mrbayes"	Path to MrBayes binary

Visualization

Parameter	Default	Description
tree_visualization_params	""	Add "--outgroup 'Species name'" to root trees; multiple outgroups: "--outgroup 'sp1,sp2'"

---

Cluster resources

Per-tool resource allocation for Slurm and PBS cluster execution. These parameters are passed automatically via the Snakemake profile — you do not need to set them manually on the command line.

Partitions / queues

Parameter	Default	Description
processing_queue	"main"	Partition for lightweight processing jobs
busco_queue	"main"	Partition for BUSCO
alignment_queue	"main"	Partition for alignment
filtration_queue	"main"	Partition for trimming
iqtree_queue	"main"	Partition for IQ-TREE
astral_queue	"main"	Partition for ASTRAL
rapidnj_queue	"main"	Partition for RapidNJ
phylip_queue	"main"	Partition for PHYLIP
raxml_queue	"main"	Partition for RAxML-NG
mrbayes_queue	"main"	Partition for MrBayes

Threads

Parameter	Default
processing_threads	1
busco_threads	8
mafft_threads	1
muscle_threads	1
prank_threads	1
clipkit_threads	1
trimal_threads	1
gblocks_threads	1
iqtree_threads	8
iqtree_per_fna_threads	1
astral_threads	4
rapidnj_threads	4
phylip_threads	1
raxml_threads	4
mrbayes_threads	8

Memory (MB)

Parameter	Default
processing_mem_mb	2000
busco_mem_mb	10000
mafft_mem_mb	2000
muscle_mem_mb	2000
prank_mem_mb	2000
clipkit_mem_mb	2000
trimal_mem_mb	2000
gblocks_mem_mb	2000
iqtree_mem_mb	10000
iqtree_per_fna_mem_mb	2000
astral_mem_mb	10000
rapidnj_mem_mb	8000
phylip_mem_mb	4000
raxml_mem_mb	10000
mrbayes_mem_mb	10000

Runtime

Parameter	Default
processing_time	"5h"
busco_time	"150h"
mafft_time	"10h"
muscle_time	"10h"
prank_time	"100h"
clipkit_time	"10h"
trimal_time	"10h"
gblocks_time	"10h"
iqtree_time	"100h"
iqtree_per_fna_time	"100h"
astral_time	"50h"
rapidnj_time	"150h"
phylip_time	"50h"
raxml_time	"100h"
mrbayes_time	"100h"