Skip to content

Configuration Reference

This page describes all parameters in the config.yaml file. The pipeline uses Snakemake's schema validation to ensure your configuration is correct before starting.

1. Core Configuration

The most commonly modified settings for any 3t-seq run. Use this annotated YAML as a guide for your basic setup.

globals:
  results_folder: "results/my_analysis/" # (1)

genome:
  label: "mm10" # (2)

sequencing_libraries:
  - name: "GSE123456" # (3)
    protocol: "pe" # (4)
    sample_sheet: "samples.csv" # (5)

defaults: # (6)
  strandedness: 0
  1. results_folder: Required. Root directory where all BAMs, counts, and logs will be saved. Always use a unique name for each experiment.
  2. label: Required. Genome version (e.g., mm10, mm39). This triggers automated downloads of all necessary references.
  3. name: Required. Unique identifier for the library or series (e.g., GEO accession).
  4. protocol: Sequencing geometry (se for single-end, pe for paired-end).
  5. sample_sheet: Path to the CSV mapping raw files to biological sample names.
  6. defaults: Global parameters that apply to all libraries unless overridden individually.

2. Library & Alignment Settings

Use these tabs to configure specialized alignment logic and library-specific overrides.

Advanced control for custom genome and which chromosomes to process.

genome:
  label: "mm10"
  fasta_path: "refs/genome.fa"
  selected_chromosomes: ["chr1", "chr2"]
fasta_path
Type: string | Default: Derived from label
Local path to a custom genome FASTA file. Required if using a non-standard reference.
gtf_path
Type: string | Default: Derived from label
Local path to a custom genome GTF annotation.
annotation_type
Type: string | Default: ensembl
Format of your GTF (ensembl, gencode, or mgi).
selected_chromosomes
Type: array | Default: null (All)
Optional list of chromosomes (e.g., ["chr1", "chr2"]). Useful for focusing analysis or speeding up pilot runs.

Custom parameters for each library in your experiment. You can customize trimming, alignment, bigwig, and other parameters for each library.

sequencing_libraries:
  - name: "GSE123"
    protocol: "pe"
    trimmomatic:
      adaptive: true
    star: "--seedSearchStartLmax 30"
trimmomatic
Type: object/string | Default: Standard flags
Custom trimming parameters. Set to adaptive: true to enable automated parameter derivation based on FastQC results.
star
Type: string | Default: ""
Extra CLI flags to pass directly to the STAR aligner for primary genome mapping.
bamCoverage
Type: string | Default: ""
Custom flags for deeptools bamCoverage (e.g., --normalizeUsing CPM).

Shared settings for statistical comparisons.

defaults:
  strandedness: 2
  deseq2:
    test: "Wald"
    variable: "condition"
deseq2
Type: object | Required: for DE
Configuration for group comparisons (Wald/LRT tests and reference levels).
strandedness
Type: integer | Default: 0 (Unstranded)
Library preparation geometry. 1: Forward, 2: Reversely stranded.

3. Expert Parameters & Module Flags

Advanced settings for internal modules. These are hidden by default to prioritize scannability.

Specialized Analysis Modules (Flags)

disable_TE_analysis: true
disable_tRNA_analysis: false
Set these to true to skip specific parts of the pipeline and reduce computation time.

  • disable_TE_analysis: Skips STAR-TE and SalmonTE quantification.
  • disable_salmonTE_analysis: Skips secondary SalmonTE processing.
  • disable_tRNA_analysis: Skips specialized tRNA mapping.
STAR-TE Internal Modes 

defaults:
  starTE_random:
    outFilterMultimapNmax: 5000
Fine-grained control over multi-mapping read assignment (multi-hits) in TEs.

  • starTE_random: Settings for multimap assignment (e.g., outFilterMultimapNmax: 5000).
  • starTE_multihit: Settings for fractional counting mode.
tRNA Expert Settings 

tRNA_quantification:
  method: "mim-tRNA-seq"
  mimseq_params:
    max_mismatches: 2
Configuration for specialized tRNA sequencing protocols.

  • method: standard or mim-tRNA-seq (for clinical-grade tRNA kits).
  • mimseq_params: Sub-parameters like max_mismatches and min_cov for the mim-tRNA-seq logic.