pandora_lib_promethion/pandora-config.example.toml

# Pandora configuration template

#######################################
# General filesystem layout / I/O
#######################################

# Directory where POD / run description files are located.
pod_dir = "/data/run_data"

# Root directory where all results will be written.
result_dir = "/mnt/beegfs02/scratch/t_steimle/data/wgs"

# Temporary directory used when unarchiving input data.
unarchive_tmp_dir = "/data/unarchived"

# Maximum memory available for dockerized tools, in GiB.
docker_max_memory_go = 400

# Path to the SQLite database of processed cases.
db_cases_path = "/data/cases.sqlite"

# Path to the conda activation script.
conda_sh = "/data/miniconda3/etc/profile.d/conda.sh"


#######################################
# Reference genome & annotations
#######################################

# Reference FASTA used throughout the pipeline.
reference = "/data/ref/hs1/chm13v2.0.fa"

# Short reference name used in filenames.
reference_name = "hs1"

# Sequence dictionary (.dict) for the reference.
dict_file = "/data/ref/hs1/chm13v2.0.dict"

# RefSeq GFF3 annotation (sorted/indexed).
refseq_gff = "/data/ref/hs1/chm13v2.0_RefSeq_Liftoff_v5.1_sorted.gff3.gz"

# Template for mask BED file (low-quality / filtered regions).
# {result_dir} -> global result directory
# {id}         -> case identifier
mask_bed = "{result_dir}/{id}/diag/mask.bed"

# BED file with early-replicating regions.
early_bed = "/data/ref/hs1/replication_early_25_hs1.bed"

# BED file with late-replicating regions.
late_bed = "/data/ref/hs1/replication_late_75_hs1.bed"

# BED file with CpG coordinates.
cpg_bed = "/data/ref/hs1/hs1/hs1_CpG.bed"

# Panels of interest: [ [name, bed_path], ... ]
panels = [
  ["OncoT",         "/data/ref/hs1/V1_V2_V3_V4_V5_intersect_targets_hs1_uniq.bed"],
  ["variable_chips","/data/ref/hs1/top_1500_sd_pos.bed"],
]


#######################################
# Sample naming / BAM handling
#######################################

# Tumor sample label (used in paths & filenames).
tumoral_name = "diag"

# Normal sample label.
normal_name = "norm"

# BAM tag name used for haplotagged reads.
haplotagged_bam_tag_name = "HP"

# Minimum MAPQ for reads kept during BAM filtering.
bam_min_mapq = 40

# Threads for BAM-level operations (view/sort/index…).
bam_n_threads = 150

# Number of reads sampled for BAM composition estimation.
bam_composition_sample_size = 20000


#######################################
# Coverage counting / somatic-scan
#######################################

# Name of directory (under each sample dir) where counts are stored.
count_dir_name = "counts"

# Bin size (bp) for count files.
count_bin_size = 1000

# Number of chunks used to split contigs for counting.
count_n_chunks = 1000

# Force recomputation of counting even if outputs exist.
somatic_scan_force = false


#######################################
# Somatic pipeline global settings
#######################################

# Force recomputation of the entire somatic pipeline.
somatic_pipe_force = true

# Default thread count for heavy tools.
somatic_pipe_threads = 150

# Template for somatic pipeline statistics directory.
# {result_dir}, {id}
somatic_pipe_stats = "{result_dir}/{id}/diag/somatic_pipe_stats"


#######################################
# Filtering / QC thresholds
#######################################

# Minimum depth in constitutional sample to consider site evaluable.
somatic_min_constit_depth = 5

# Maximum allowed ALT count in constitutional sample for a somatic call.
somatic_max_alt_constit = 1

# Window size (bp) for sequence entropy around variants.
entropy_seq_len = 10

# Minimum Shannon entropy threshold.
min_shannon_entropy = 1.0

# Max depth considered "low quality".
max_depth_low_quality = 20

# Min depth considered "high quality".
min_high_quality_depth = 14

# Minimum number of callers required to keep a variant.
min_n_callers = 1


#######################################
# DeepVariant configuration
#######################################

# DeepVariant output directory template.
# {result_dir}, {id}, {time}
deepvariant_output_dir = "{result_dir}/{id}/{time}/DeepVariant"

# Threads for DeepVariant.
deepvariant_threads = 150

# DeepVariant version / image tag.
deepvariant_bin_version = "1.9.0"

# DeepVariant model type (e.g. ONT_R104).
deepvariant_model_type = "ONT_R104"

# Force DeepVariant recomputation.
deepvariant_force = false


#######################################
# DeepSomatic configuration
#######################################

# DeepSomatic output directory template.
# {result_dir}, {id}, {time}
deepsomatic_output_dir = "{result_dir}/{id}/{time}/DeepSomatic"

# Threads for DeepSomatic.
deepsomatic_threads = 150

# DeepSomatic version / image tag.
deepsomatic_bin_version = "1.9.0"

# DeepSomatic model type.
deepsomatic_model_type = "ONT"

# Force DeepSomatic recomputation.
deepsomatic_force = false


#######################################
# ClairS configuration
#######################################

# Threads for ClairS.
clairs_threads = 155

# ClairS docker tag.
clairs_docker_tag = "latest"

# Force ClairS recomputation.
clairs_force = false

# Platform preset for ClairS.
clairs_platform = "ont_r10_dorado_sup_5khz_ssrs"

# ClairS output directory template.
# {result_dir}, {id}
clairs_output_dir = "{result_dir}/{id}/diag/ClairS"


#######################################
# Savana configuration
#######################################

# Savana binary (name or full path).
savana_bin = "savana"

# Threads for Savana.
savana_threads = 150

# Savana output directory template.
# {result_dir}, {id}
savana_output_dir = "{result_dir}/{id}/diag/savana"

# Savana copy-number output file.
# {output_dir}, {id}, {reference_name}, {haplotagged_bam_tag_name}
savana_copy_number = "{output_dir}/{id}_diag_{reference_name}_{haplotagged_bam_tag_name}_segmented_absolute_copy_number.tsv"

# Savana raw read counts file.
savana_read_counts = "{output_dir}/{id}_diag_{reference_name}_{haplotagged_bam_tag_name}_raw_read_counts.tsv"

# Savana passed VCF.
savana_passed_vcf = "{output_dir}/{id}_diag_savana_PASSED.vcf.gz"

# Force Savana recomputation.
savana_force = false

# Constitutional phased VCF template.
# {result_dir}, {id}
germline_phased_vcf = "{result_dir}/{id}/diag/{id}_variants_constit_phased.vcf.gz"


#######################################
# Severus configuration
#######################################

# Path to Severus script.
severus_bin = "/data/tools/MySeverus/severus.py"

# Force Severus recomputation.
severus_force = false

# Threads for Severus.
severus_threads = 32

# VNTRs BED for Severus.
vntrs_bed = "/data/ref/hs1/vntrs_chm13.bed"

# Path of the Severus panel of normals.
severus_pon = "/data/ref/hs1/PoN_1000G_chm13.tsv.gz"

# Paired Severus output directory.
# {result_dir}, {id}
severus_output_dir = "{result_dir}/{id}/diag/severus"

# Solo Severus output directory.
# {result_dir}, {id}, {time}
severus_solo_output_dir = "{result_dir}/{id}/{time}/severus"


#######################################
# Longphase configuration
#######################################

# Path to longphase binary.
longphase_bin = "/data/tools/longphase_linux-x64"

# Threads for longphase.
longphase_threads = 150

# Threads for longphase modcall step.
longphase_modcall_threads = 8

# Longphase modcall VCF template.
# {result_dir}, {id}, {time}
longphase_modcall_vcf = "{result_dir}/{id}/{time}/5mC_5hmC/{id}_{time}_5mC_5hmC_modcall.vcf.gz"


#######################################
# Modkit configuration
#######################################

# Path to modkit binary.
modkit_bin = "modkit"

# Threads for `modkit summary`.
modkit_summary_threads = 50

# Modkit summary file template.
# {result_dir}, {id}, {time}
modkit_summary_file = "{result_dir}/{id}/{time}/{id}_{time}_5mC_5hmC_summary.txt"


#######################################
# Nanomonsv configuration
#######################################

# Path to nanomonsv binary.
nanomonsv_bin = "/home/prom/.local/bin/nanomonsv"

# Paired nanomonsv output directory template.
# {result_dir}, {id}, {time}
nanomonsv_output_dir = "{result_dir}/{id}/{time}/nanomonsv"

# Force nanomonsv recomputation.
nanomonsv_force = false

# Threads for nanomonsv.
nanomonsv_threads = 150

# Paired nanomonsv PASSED VCF template.
# {output_dir}, {id}
nanomonsv_passed_vcf = "{output_dir}/{id}_diag_nanomonsv_PASSED.vcf.gz"

# Solo nanomonsv output directory template.
# {result_dir}, {id}, {time}
nanomonsv_solo_output_dir = "{result_dir}/{id}/{time}/nanomonsv-solo"

# Solo nanomonsv PASSED VCF template.
# {output_dir}, {id}, {time}
nanomonsv_solo_passed_vcf = "{output_dir}/{id}_{time}_nanomonsv-solo_PASSED.vcf.gz"


#######################################
# PromethION metadata
#######################################

# Directory containing PromethION run metadata.
promethion_runs_metadata_dir = "/data/promethion-runs-metadata"

# JSON file mapping flowcell IDs / runs for Pandora.
promethion_runs_input = "/data/pandora-flowcell-id.json"


#######################################
# Alignment / basecalling (Dorado)
#######################################

[align]
# Path to Dorado binary.
dorado_bin = "/mnt/beegfs02/scratch/t_steimle/tools/dorado-latest-linux-x64/bin/dorado"

# Dorado basecalling arguments (device, model, modifications…).
dorado_basecall_arg = "-x 'cuda:all' sup,5mC_5hmC"

# Should dorado re-align after demux ?
dorado_should_realign = false

# Dorado aligner threads number
dorado_aligner_threads = 20

# Reference FASTA used for alignment.
ref_fa = "/mnt/beegfs02/scratch/t_steimle/ref/hs1/chm13v2.0.fa"

# Minimap2 index used for alignment.
ref_mmi = "/mnt/beegfs02/scratch/t_steimle/ref/chm13v2.0.mmi"

# Samtools bin 
samtools_bin = "/mnt/beegfs02/scratch/t_steimle/tools/samtools"

# Threads for `samtools view`.
samtools_view_threads = 20

# Threads for `samtools sort`.
samtools_sort_threads = 48

# Threads for `samtools merge`.
samtools_merge_threads = 48

# Threads for `samtools split`.
samtools_split_threads = 48