DE_config.yaml

# ---------------------- QUICK START ----------------------------------
# 1. Edit the organism defaults file (and, optionally, change the settings 
#    under `organism:`)
# 2. Configure the in_path_pattern under `pipeline:`. This depends on the
#    location of the files with the count data.
# 3. Run `sea-snap covariate_file star` or 
#    `sea-snap covariate_file salmon` to automatically generate the covariate file.
#    Remember to edit it.
# 4. Edit the `experiment:` section:
#    - add the design formula (e.g. "~ group")
#    - for each variable used in formula, add an entry in "columns:"
#      listing the levels of that variable
# 5. Go to the `report:` section and choose which snippets you wish to run.
# 6. Test the configuration with `sea-snap DE l -np`
# 7. Run the pipeline locally (e.g. `sea-snap DE l --cores 4`), or
#    on the cluster (`sea-snap DE --slurm c`).


# 1. --- organism annotation

organism_defaults: ### FILL HERE ###  (e.g. Homo_sapiens.yaml) 

organism:
  __options__: [name, genus, taxon, files, star_index, salmon_index, R]
  ### OVERWRITE ###  organism defaults (e.g. gtf, genome and indices)

# 2. --- general pipeline parameters

pipeline_param:
  __options__: [out_path_pattern, log_path_pattern, in_path_pattern, report_snippets, input_choice]

  in_path_pattern: ### FILL HERE ###
  out_path_pattern: DE/{contrast}/{step}/out/{step}.{contrast}.{extension}
  log_path_pattern: DE/{contrast}/{step}/report/{step}.{contrast}.{extension}
  ## Notes:
  ## The in_path_pattern can be the same as the out_path_pattern of the mapping pipeline
  ## You should be able to safely leave the predefined {out,log}_path_pattern[s].

  # define choices for choose_input()
  # select the one that suits you
  input_choice:
    mapping:
      - import_gene_counts
      - import_sf
      - import_featurecounts



#--- information about experiment

experiment:
  __options__: [covariate_file, design_formula, columns]
  
  design_formula: ### FILL HERE ###  (e.g. "~ group")

## the following allows you define the orders of levels in the factor
## variables
  columns: ### FILL HERE ###
## example:
#  columns:
#    treatment:
#      - control
#      - treatment
#    strain:
#      - WT
#      - KO

    
#--- creation of contrasts

contrasts:
  __options__: [defaults, contrast_list]

  ## by default: run tmod and cluster_profiler, but not tmod_pca
  defaults:
    tmod: true 
    tmod_pca: false
    cluster_profiler:
      run: true
  
  contrast_list:
    - title: ### FILL HERE ###  (unique title)
      ### One of coef, ratio or vector must be set for each contrast
      coef:
      ratio:
        column:
        numerator:
        denominator:
      vector:


#--- filtering of count matrices

filters:
  __options__: [low_counts, experiment_blacklist, experiment_whitelist]

  # at least 10 counts per row
  low_counts: 10

  # these two parameters allow a more fine grained filtering:
  # setting them means that only genes will be kept which have at least
  # `min_counts` in at least `min_count_n` samples

  min_counts: 5
  min_count_n: 3

#--- normalization parameters

normalization:
  __options__: [DESeq2_parameters]


#--- QC parameters

QC:
  __options__: [countMatrix_heatmap, sample_similarity, samples_pca]


#--- report parameters

report:
  __options__: [defaults, report_snippets, snippet_parameters]

  ## to choose other snippet parameters,
  ## take a look at sea-snap/defaults/DE_config_defaults.yaml

  ## Additional directory with RMD snippets
  path: "snippets"
 
  defaults:
    ## this is a special construct
    ## to add the snippets for each
    ## of the contrasts
    contrast:
      - __list__: __contrasts__
    contrast_list:
      - Init_code.Rmd
      - MA_plot.Rmd
      - Result_table.Rmd
      - Goseq_GO_table.Rmd
      - tmod_contrast.Rmd
      - cluster_profiler: __defaults__
    cluster_profiler:
      - Dot_plot.Rmd
      - EnrichmentMap_plot.Rmd
      - Upset_plot.Rmd

  ## comment snippets which you do not want to include
  report_snippets:
    - Covariate_table.Rmd
    - NormalisationQC: 
      - PCA_plot.Rmd
      - HierarchicalClustering_plot.Rmd
      - SampleSimilarity_plot.Rmd
      - NormExprDens_plot.Rmd
      - DESeq2Norm_plot.Rmd
      - CoxOutlier_plot.Rmd
    # volcano plot for each contrast
    - Volcano.Rmd
    - contrast: __defaults__
    # a summary of enrichment results for all contrasts
    # in one place
    - Functional:
      - tmod.Rmd
      - cluster_profiler_summary.Rmd
      # - Disco.Rmd
      
# ---------------------------------------- configuration for tmod
# the following defines the gene set enrichment gene sets
tmod:
  tmod_db_path: "./"
  sort_by: pval 
# for sorting order, you can also use pval_n and/or pval_p
# (pval negative sort, pval positive sort)
  databases:
    - name: tmod
      file: tmod
      title: "Co-expression gene sets (tmod)"
      description:
        Gene sets derived from clustering expression profiles from human blood
        collected for various immune conditions. These gene sets are included
        in the tmod package by default. Check tmod documentation for further 
        information.
      taxonID: 9606
      primaryID: SYMBOL
      annotationDBI: org.Hs.eg.db
    - name: msigdb_reactome
      file: msigdb
      title: "Reactome gene sets (MSigDB)"
      description:
        Reactome gene sets
        the Molecular Signatures DB 
        (https://www.gsea-msigdb.org/gsea/msigdb/).
      primaryID: ENTREZID
      subset: "Category=C2,Subcategory=CP:REACTOME"
    - name: msigdb_hallmark
      file: msigdb
      title: "Hallmark gene sets (MSigDB)"
      description:
        Hallmark gene sets
        the Molecular Signatures DB 
        (https://www.gsea-msigdb.org/gsea/msigdb/).
      primaryID: ENTREZID
      subset: "Category=H"
    - name: msigdb_kegg
      file: msigdb
      title: "KEGG pathways (MSigDB)"
      description:
        KEGG pathways from 
        the Molecular Signatures DB 
        (https://www.gsea-msigdb.org/gsea/msigdb/).
      primaryID: ENTREZID
      subset: "Category=C2,Subcategory=CP:KEGG"
    - name: msigdb_mir
      file: msigdb
      title: "MIR targets (MSigDB)"
      description:
        MIR targets from
        the Molecular Signatures DB 
        (https://www.gsea-msigdb.org/gsea/msigdb/).
      primaryID: ENTREZID
      subset: "Category=C3,Subcategory=MIR:MIRDB"
    - name: msigdb_go_bp
      file: msigdb
      title: "GO Biological Process (MSigDB)"
      description:
        GO Biological Process definitions from
        the Molecular Signatures DB 
        (https://www.gsea-msigdb.org/gsea/msigdb/).
      primaryID: ENTREZID
      subset: "Category=C5,Subcategory=GO:BP"