nextflow.config.Visium.mouse

// Default configs

params {

  // General
  // No need to modify
  workflowpath = "${projectDir}"
  // author name to add to the final report
  authorname = "Liu,Leo"
  // one of scRNAseq, Visium
  data_type = "Visium"
  // Full path to a .txt file containing a list of genes,one gene per line
  // set to "NA" to disable
  feature_list = "${projectDir}/testdata/feature_gene.txt"
  // output directory
  output_dir = "${projectDir}/output/"
  // Path to gene level annotation file. This is used to add feature level meta data
  geneinfo = "${projectDir}/docs/mouse_gene_info_2020A.tsv"
  
  // QC
  qc_only = true
  // "0" or "1" to indicate whether to perform ambient RNA removal/correction using SoupX
  // only applicable to scRNAseq
  ambient_RNA_removal_flag = "0"
  // "0" or "1" to indicate whether to perfrom doublet removal using scDblFinder
  // only applicable to scRNAseq  
  doublet_removal_flag = "0"
  // "0" or "1" to indicate whether to apply adaptive cutoff idenfication (based on IQR)
  adaptive_cutoff_flag = "1"
  // Cutoff for percentage of mitochondria concentration
  // cells with values higher than the cutoff will be removed
  mt_cutoff = 40
  // Cutoff for percentage of hemoglobin concentration
  // cells with values higher than the cutoff will be removed
  hb_cutoff = 20
  // Cutoff for total number of detectable genes/features
  // cells with values lower than the cutoff will be removed
  nFeature_cutoff = 50
  // Cutoff for total number of UMI counts
  // cells with values lower than the cutoff will be removed
  nCount_cutoff = 200
  // Cutoff for number of cells with expression for feature/gene
  // genes/features with values lower than the cutoff will be removed
  nCell_cutoff = 10

  // Normalization
  // normalization method for dimension reduction and clustering, either SCT or LogNormalize
  norm_dimreduc = "LogNormalize"
  // normalization method for differential testing, either SCT or LogNormalize
  norm_diff = "LogNormalize"
  // cell-cyle 
  // "0" or "1" to indicate whether to estimate and correct for cell-cycle effect.
  cellcycle_correction_flag = "1"
  // path to gene list (Gene symbols) for cell-cycle S-phase, one gene per line
  // required if cellcycle_correction is set to "1"
  genelist_S_phase = "${projectDir}/docs/S_genes_mouse.tsv"
  // path to gene list (gene symbols) for cell-cycle G2M-phase, one gene per line
  // required if cellcycle_correction is set to "1"
  genelist_G2M_phase = "${projectDir}/docs/G2M_genes_mouse.tsv"

  // Analysis strategy
  // "0" or "1" indicating whether to perform merge-based analysis
  merge_analysis = false
  // "0" or "1" indicating whether to perform integration-based analysis
  integration_analysis = true
  // if merge_analysis is enabled, whether to stop after merge-based analysis. Could be useful if you want to evaluate parameters like resolution.
  merge_only = false
  // if integration_analysis is enabled, whether to stop after integration-based analysis. Could be useful if you want to evaluate parameters like resolution.
  integration_only = false

  // Integration strategy
  // cca, rpca, harmony, fastmnn, or scvi
  integration_method = "harmony"
  // "0" or "1" to indicate whether to perform sketch-based workflow
  sketch_flag = "0"

  // Clustering
  // Resolution parameter used to identify number of clusters
  resolution = 0.8
  // Method used for visualization, either tsne or umap
  vismethod = "umap"
  // method for spatial clustering, Banksy or Seurat
  // only applicable to spatial data
  spatial_cluster = "Banksy"
  // lambda parameter for Banksy, larger values yield more spatially coherent domains. For Visuim, recommend to use 0.2 for domain segmentation. For Visuim HD, use 0.2 for cell typing, and 0.8 for domain segmentation
  // only applicable to spatial data
  lambda = 0.2
  // k_geom parameter for Banksy, larger values will yield larger domains. Recommend to use 18 for Visium data
  // only applicable to spatial data
  k_geom = 18

  // Differential expression
  // Character value specifying control group for differential expression analysis
  // Set to NA to disable
  control_var = "Control"
  // Character value specifying case group for differential expression analysis
  // Set to NA to disable
  case_var = "Case"
  // Covariates to adjust, when performing differential analysis between conditions
  // values should be from column names from sampleinfo file
  // e.g, if sampleinfo file contains information for 'age' and 'gender',
  // and you want to correct for age and gender effects, you can specify
  // covariate_list = "age,gender"
  // note that if covariate_list is set, you will need to choose one of 
  // LR, negbinom, poisson or MAST as method for statistical test. 
  covariate_list = "NA"
  // Denotes which statistical test to use
  // refer to FindMarkers function for more details
  test = "wilcox"
  // fold change cutoff to identify differentially expressed genes
  fc = 2
  // p value cutoff to identify differentially expressed genes
  pval = 0.01
  // "0" or "1" to indicate whether to use Bonferroni adjusted p value
  pval_flag = "1"
  // percentage of experssion cutoff to identify differentially expressed genes
  // require at least one of the group (e.g. control or case)
  // to have percentage of cells expressing a gene above specified value.
  pct=20
}


profiles {
    // Default profile (local execution)
    local {
      process.executor = 'local'
      process.memory = '100Gb'
      process.INTEGRATESAMPLES.memory = '200G'
      workDir = './work'
    }     

    // SLURM execution profile
    slurm {
      process.executor = 'slurm'
      process.queue = 'cpu-short'
      //process.memory = '100Gb'
      //process.INTEGRATESAMPLES.memory = '200G'
      process.clusterOptions = '--cpus-per-task 10  --mem 200G'
      process.time = '6h'
      workDir = './work'
    }
}

manifest {
    name = 'STITCH'
    author = 'Yuanhang Liu'
    description = 'A nextflow pipeline for scRNA-seq and Visium data analysis'
    version = '1.0.0'
}