Haplotagging (Short-read DNA)

Phase small variants in short-read DNA sequencing BAM files

usage

nexus run --nf-workflow haplotagging_short-read-dna.nf -params-file params.yaml

Note

Nextflow config files are available here. Use the config file that matches your installed nexus version (e.g. nexus_v0.2.0_nextflow_slurm.config).

parameters

Download params.yaml

# =============================================================================
# params.yaml — haplotagging_short-read-dna
#
# Usage:
#   nextflow run haplotagging_short-read-dna.nf -params-file params.yaml
#
# Two orthogonal selectors decide what runs (CSV list, "all", or "none"/""):
#   small_variant_callers   {deepvariant, haplotypecaller, clair3,
#                            strelka2-germline}
#   phasing_methods         {whatshap, hapcut2-whatshap}
#
# Each requested phaser runs ONCE PER selected small-variant caller, with
# outputs at ${output_dir}/${sample_id}/<caller>_<phaser>/.
#
# NOTE: HapCUT2 requires strict diploid GTs (alleles in {0,1,2}, no '.'
# calls). WGS DeepVariant VCFs may contain non-diploid GTs — if a HapCUT2
# task fails on 'Non-diploid VCF entry detected', drop "hapcut2-whatshap"
# from phasing_methods or pre-filter the VCF.
# =============================================================================


# ---- required: input/output paths -------------------------------------------

# TSV file with columns: sample_id, bam_file, bam_bai_file
samples_tsv_file: ""

# Directory to which output files will be copied
output_dir: ""

# Reference genome FASTA file (may be .fa or .fa.gz)
reference_genome_fasta_file: ""


# ---- required: pipeline selectors -------------------------------------------

small_variant_callers: "all"
phasing_methods: "all"

# Output format for haplotagged reads.
#   "bam"   →  publish full haplotagged BAM(s) (default).
#   "tsv"   →  publish only a haplotag TSV per phaser (~1000× smaller).
#   "both"  →  publish BAM and TSV.
# Phased VCFs are always published.
haplotag_output: "bam"


# =============================================================================
# Tool-specific settings (only consumed when the corresponding tool runs).
# =============================================================================

# DeepVariant — input_path / output_path are host paths bind-mounted into the
# container. Both REQUIRED when running DeepVariant.
deepvariant:
  input_path:  ""
  output_path: ""
  containerization: "singularity"   # "singularity" | "docker"
  model_type: "WGS"                 # WGS | WES (use WGS for whole-genome short-read)
  bin_path: "/opt/deepvariant/bin/run_deepvariant"
  bin_version: "1.9.0"

# GATK4 HaplotypeCaller — one invocation per chromosome, merged via Picard MergeVcfs.
haplotypecaller:
  extra_args: ""
  chromosomes: "chr1,chr2,chr3,chr4,chr5,chr6,chr7,chr8,chr9,chr10,chr11,chr12,chr13,chr14,chr15,chr16,chr17,chr18,chr19,chr20,chr21,chr22,chrX,chrY,chrM"

# Clair3 — default extra_args use the Illumina model bundled in the image.
clair3:
  extra_args: "--platform=ilmn --include_all_ctgs"

# Strelka2-germline
strelka2_germline:
  extra_args: ""

# WhatsHap (also used as fallback for hapcut2_whatshap.whatshap_haplotag_extra_args)
whatshap:
  phase_extra_args: "--mapq 20"
  haplotag_extra_args: "--ignore-read-groups --skip-missing-contigs --output-threads 4"

# HapCUT2 + WhatsHap haplotag — see note above on diploid GT requirements.
hapcut2_whatshap:
  read_technology: "illumina"        # "pacbio" | "ont" | "illumina"
  extracthairs_extra_args: ""
  hapcut2_extra_args: ""
  whatshap_haplotag_extra_args: "--ignore-read-groups --skip-missing-contigs --output-threads 4"