Skip to content
Snippets Groups Projects
Code owners
Assign users and groups as approvers for specific file changes. Learn more.
config.yaml 3.09 KiB
#User settings:

data:
  samples: "" #path to the samples-file
  units: "" #path to the units-file

general:
  amount_of_hvgs: "" #amount of highly variable genes used in the analysis

preprocessing:
  path_to_bamtofastq: "" #path to the bamtofastq program 
  path_to_cellranger: "" #path to the cellranger program
  path_to_ref: "" #path to the reference genome used


quality_control:
  cells: #filter of cells
    upper_quantile_counts: "0.95"  #maximum counts per cell; use one parameter, leave the other as ""
    max_counts: ""
 
    lower_quantile_counts: "0.05"  #minimum counts per cell; use one paramter, leave the other as "" 
    min_counts: ""
 
    lower_quantile_genes: "0.05" #minimum amount of genes per cell; use one, leave the other as ""
    min_genes: ""
 
    mt_frac: "" #maximum proportion of mitochondrial genes per cell

  genes:
    min_cells: "" #minimum amount of cells a gene has to be expressed in

downsampling:
  downsampling_method: "normal" #options: "sphetcher" or "normal"; "normal" =  no downsampling
  path_to_sphetcher: "" #path to sphetchers src-folder (only used if downsampling_method = "sphetcher")
  sketch_size: "" #amount of cells after downsampling (only used if downsampling_method = "sphetcher")

cell_cycle_scoring:
  ref_genes: "" #used for visualization of cell cycle effects; tsv-file with 2 columns(s and g2m) defining cell cycle genes; OPTIONAL 

clustering:
  celltypes_markergenes: "" #known celltypes with their marker genes; tsv-file with 2 columns(celltype and markergenes), celltype definies a specific celltype & markergenes its markergenes as a list
  genes_to_vis: ""   #genes that are visualized across the clustering; OPTIONAL
  subclustering: ""    #options: "" or "True"; "" = no automatic subclustering
  clustering_resolution: "0.5" #sensitivity parameter for the louvain-algorithm
  cluster_method: "louvain" #options: "specter" or "louvain"; used algorithm for clustering

specter: #parameters only used if clustering:cluster_method = "specter"
  number_of_clusters: "8" #number of clusters in the final clustering
  ensemble_size: "200" 
  mingamma: "0.1"

subclustering:
  subclustering_resolution: "0.2" #sensitivity parameter for the lovain-algorithm
  names_for_unannotated: "" #names to use for the unannotated clusters; OPTIONAL
  further_subclusterings: "" #clusters to subcluster; OPTIONAL
 
trajectory_inference:
  clusters_to_include: "" #clusters that are used for trajectory inference; all clusters used if parameter is "" 
  trajectory_start: "" #cluster thats the starting point of the trajectory; random cluster chosen if parameter is ""
  trajectory_ending: "" #cluster thats the end point of the trajectory; random cluster chosen if parameter is ""

differential_testing:
  clusters: "" #clusters used for differential testing, if a subcluster is supposed to be tested: write cluster-0 instead of cluster,0; all clusters used if parameter is ""
  DE_threshold: "0.01" #threshold for p-values of differentially expressed genes 

gene_set_enrichment_analysis:
  enrichment_threshold: "0.05" #threshold for p-values of enriched GO-terms  
  organism: "" #analysed organism