config.template.yaml

options:
  sleep_time: 1 #in hour
  verbosity: false

# static snakemake params that will passed
snakemake:                                                                                                
  conda_prefix: /path/to/conda/prefix
  cores: 4
  dryrun: False
  printdag: False
  max_jobs_per_second: 1
  printshellcmds: true
  use_conda: true
  verbose: true
  rerun_triggers: ['mtime']
  debug: False

# ssh credential for samba host
sambahost:
  user: samba_user
  host: samba_host
  pkey: path/to/private_key

paths:
  offloadDir: /path/to/ont/files (fast5 or pod5)      (output from ONT)
  old_outputDirs:
    - /path/to/output/old_for_legacy
    - /path/to/output/old_for_legacy2
  outputDir: /path/to/output/the_real_one             (aka as work dir)
  groupDir: /rootpath/to/transfer/for/internal/groups (aka "periphery")
  external_groupDir: /rootpath/to/transfer/for/external/groups
  deepseq_qc: /path/to/deepseq/qc/ (SAMBA)
  bioinfocoredir: /path/to/bionfo_qc (scp)
  rulesDir: relative_path_to_snakemake_rules # e.g. rules_dorado ('rules' is deprecated)
ignore:
  flowcells:
    # flowcell IDs to ignore in the offload_path
    - 'PAK83895'
    - ...
  dirs:
    # flowcell paths to ignore
    - '20230512_1047_P2-7085206e51685c-A_PAK73534_a5e9ed5e'
    - ...

# credential to query parkour
parkour:
  url: parkour.domain.tld  # leave empty to disable Parkour access
  user: account
  password: password
  pem: path/to/certificate


basecaller: dorado

dorado_basecaller:
  dorado_cmd: /path/to/dorado/basecaller
  dorado_version: 
  dorado_options: --recursive
  dorado_modifications: --modified-bases 5mCG_5hmCG
  dorado_output: fastq # set to activate bam2fastq conversion (default for legacy)
  dorado_model: # model name to overwrite inferred name
  model_directory: /path/to/dorado/models

#fastqc:
#  memory: 10000

kraken:
  db: /path/to/kraken2/db

multiqc:
  configfile: /path/to/multiqc/config/file

# organism labels, as set on Parkour
genome:
  drosophila: /path/to/genome.fa
  mouse: /path/to/genome.fa
  human: /path/to/genome.fa
  lambdaPhage: /path/to/genome.fa
  human_rRNA: /path/to/genome.fa
  mouse_rRNA: /path/to/genome.fa

# organism labels, as set on Parkour
transcripts:
  drosophila:  /path/to/genes.bed
  mouse: /path/to/genes.bed
  human: /path/to/genes.bed

contamination_report:
human_genome: /path/to/genome.fa
human_rRNA: /path/to/genome.fa
mouse_genome: /path/to/genome.fa
mouse_rRNA: /path/to/genome.fa

# options to skip heavy computations
default_process:
  do_basecall: 'do_basecall'
  do_align: 'do_align'
  do_modbed: 'do_modbed'

bam_merge:
  batch_size: 500
  opt: '-c --no-PG'

porechop:
  sample_reads: 100000 # as porechop is slow, this is the max number of reads per sample to analyze

email:
  from: a@host
  to: b@host
  trigger: c@host
  host: host