Dev nf

.github/workflows/ci.yml

@@ -41,9 +41,9 @@ jobs:
 
       - name: Build new docker image
         if: ${{ github.event.workflow_run.conclusion == 'thiswillnevertrigger' }}
-        run: docker build --no-cache . -t almurphy/scfdev:dev
+        run: docker build --no-cache . -t nfancy/scflow:dev
       - name: Pull docker image
         if: ${{ github.event.workflow_run.conclusion == 'thiswillnevertrigger' }}
         run: |
-          docker pull almurphy/scfdev:dev
-          docker tag almurphy/scfdev:dev almurphy/scfdev:dev
+          docker pull nfancy/scflow:dev
+          docker tag nfancy/scflow:dev nfancy/scflow:dev

.nf-core.yml

@@ -9,7 +9,7 @@ lint:
 
   schema_params:
     - input
-
 actions_awsfulltest: False
 pipeline_todos: False
 actions_ci: False
+repository_type: pipeline

README.md

@@ -59,7 +59,7 @@ Try the pipeline on an in-built, minimal test dataset (all inputs will be automa
 3. Download the pipeline and test it on a minimal dataset with a single command:
 
     ```console
-    nextflow run nf-core/scflow -profile test,<docker/singularity/podman/shifter/charliecloud/conda/institute>
+    nextflow run combiz/nf-core-scflow -r dev-nf -profile test,<docker/singularity/podman/shifter/charliecloud/conda/institute>
     ```
 
     > - Please check [nf-core/configs](https://github.com/nf-core/configs#documentation) to see if a custom config file to run nf-core pipelines already exists for your Institute. If so, you can simply use `-profile <institute>` in your command. This will enable either `docker` or `singularity` and set the appropriate execution settings for your local compute environment.
@@ -75,7 +75,7 @@ A complete, automated, scalable, and reproducible case-control analysis can then
 1. Start running your own analysis!
 
     ```bash
-    nextflow run nf-core/scflow \
+    nextflow run combiz/nf-core-scflow -r dev-nf \
     --manifest Manifest.tsv \
     --input Samplesheet.tsv \
     -c scflow_params.config \

bin/scflow_dge.r

@@ -131,15 +131,15 @@ required$add_argument(
 )
 
 required$add_argument(
-  "--fc_threshold",
+  "--logFC_threshold",
   type = "double",
   default = 1.1,
   metavar = "number",
   help = "Absolute fold-change cutoff for DE [default %(default)s]"
 )
 
 required$add_argument(
-  "--pval_cutoff",
+  "--padj_cutoff",
   type = "double",
   default = 0.05,
   metavar = "number",
@@ -254,8 +254,6 @@ de_results <- perform_de(
   ref_class = args$ref_class,
   confounding_vars = args$confounding_vars,
   random_effects_var = args$random_effects_var,
-  fc_threshold = args$fc_threshold,
-  pval_cutoff = args$pval_cutoff,
   mast_method = args$mast_method,
   force_run = args$force_run,
   ensembl_mapping_file = args$ensembl_mappings,
@@ -278,8 +276,8 @@ for (result in names(de_results)) {
     )
 
     report_de(de_results[[result]],
-      fc_threshold = args$fc_threshold,
-      pval_cutoff = args$pval_cutoff,
+      logFC_threshold = args$logFC_threshold,
+      padj_cutoff = args$padj_cutoff,
       n_label = args$n_label,
       report_folder_path = file.path(getwd()),
       report_file = paste0(file_name, result, "_scflow_de_report")
@@ -289,8 +287,8 @@ for (result in names(de_results)) {
 
     p <- scFlow::volcano_plot(
       dt = de_results[[result]],
-      fc_threshold = args$fc_threshold,
-      pval_cutoff = args$pval_cutoff,
+      logFC_threshold = args$logFC_threshold,
+      padj_cutoff = args$padj_cutoff,
       n_label = args$n_label
     )
 

bin/scflow_dirichlet.r

@@ -65,11 +65,28 @@ required$add_argument(
   required = TRUE
 )
 
+required$add_argument(
+  "--confounding_vars",
+  help = "confounding variables",
+  metavar = "sex",
+  required = TRUE
+)
+
 # get command line options, if help option encountered print help and exit,
 # otherwise if options not found on command line then set defaults
 args <- parser$parse_args()
-args$var_order <- strsplit(args$var_order, ",")[[1]]
-if (tolower(args$var_order) == "null") { args$var_order <- NULL }
+
+if (tolower(args$var_order) == "null") { 
+  args$var_order <- NULL 
+  } else {
+    args$var_order <- strsplit(args$var_order, ",")[[1]]
+  }
+
+if (tolower(args$confounding_vars) == "null") { 
+  args$confounding_vars <- NULL 
+  } else {
+    args$confounding_vars <- strsplit(args$confounding_vars, ",")[[1]]
+  }
 
 #   ____________________________________________________________________________
 #   Start                                                                   ####
@@ -82,7 +99,8 @@ results <- model_celltype_freqs(
   celltype_var = args$celltype_var,
   dependent_var = args$dependent_var,
   ref_class = args$ref_class,
-  var_order = args$var_order
+  var_order = args$var_order,
+  confounding_vars = args$confounding_vars
 )
 
 ## ............................................................................

bin/scflow_ipa.r

@@ -64,15 +64,15 @@ required$add_argument(
 )
 
 required$add_argument(
-  "--fc_threshold",
+  "--logFC_threshold",
   type = "double",
-  default = 1.1,
+  default = 0.5,
   metavar = "number",
   help = "Absolute fold-change cutoff for DE [default %(default)s]"
 )
 
 required$add_argument(
-  "--pval_cutoff",
+  "--padj_cutoff",
   type = "double",
   default = 0.05,
   metavar = "number",
@@ -121,8 +121,8 @@ for (gene_file in args$gene_file) {
 
   dt <- dt %>%
     dplyr::filter(
-      padj <= args$pval_cutoff,
-      abs(logFC) >= log2(args$fc_threshold)
+      padj <= args$padj_cutoff,
+      abs(logFC) >= args$logFC_threshold
     )
 
   if (nrow(dt) < 5) {
@@ -134,11 +134,32 @@ for (gene_file in args$gene_file) {
       organism = getOption("scflow_species"),
       enrichment_tool = args$enrichment_tool,
       enrichment_method = args$enrichment_method,
-      enrichment_database = args$enrichment_database,
-      is_output = TRUE,
-      output_dir = output_dir
+      enrichment_database = args$enrichment_database
     )
 
+    for(i in names(enrichment_result)){
+
+      output_dir_path <- paste(output_dir, i, sep = "/")
+      dir.create(output_dir_path)
+
+      res <- enrichment_result[[i]]
+
+      lapply(
+          setdiff(names(res), c("plot", "metadata")),
+          function(dt) {
+            write.table(res[dt],
+                        file = paste(output_dir_path, "/", dt, ".tsv", sep = ""),
+                        row.names = FALSE,
+                        col.names = gsub(dt, "", colnames(res[[dt]])), sep = "\t")})
+        lapply(
+          names(res$plot),
+          function(p) {
+            ggplot2::ggsave(paste(output_dir_path, "/", p, ".png", sep = ""),
+                            res$plot[[p]],
+                            device = "png", height = 8,
+                            width = 10, units = "in", dpi = 300)})
+    }
+
     if (all(unlist(lapply(
       enrichment_result, function(dt) {
         isFALSE(dt$metadata$result)

bin/scflow_map_celltypes.r

@@ -53,6 +53,15 @@ required$add_argument(
   required = TRUE
 )
 
+required$add_argument(
+  "--annotation_level",
+  default = 1,
+  type = "integer",
+  help = "the annotation level of the reference ctd",
+  required = TRUE,
+  metavar = "N"
+)
+
 required$add_argument(
   "--species",
   help = "the biological species (e.g. mouse, human)",
@@ -101,6 +110,7 @@ sce <- map_celltypes_sce(
   ctd_folder = args$ctd_folder,
   clusters_colname = args$clusters_colname,
   cells_to_sample = args$cells_to_sample,
+  annotation_level = as.numeric(args$annotation_level),
   species = args$species
 )
 

bin/scflow_reduce_dims.r

@@ -313,6 +313,7 @@ args <- purrr::map(args, function(x) {
 
 sce <- read_sce(args$sce_path, read_metadata = TRUE)
 
+set.seed(42)
 sce <- reduce_dims_sce(
   sce,
   input_reduced_dim = args$input_reduced_dim,

conf/base.config

@@ -40,15 +40,17 @@ process {
         time   = { check_max( 8.h   * task.attempt, 'time'    ) }
     }
     withLabel:process_high {
-        cpus   = { check_max( 12    * task.attempt, 'cpus'    ) }
+        cpus   = { check_max( 8    * task.attempt, 'cpus'    ) }
         memory = { check_max( 72.GB * task.attempt, 'memory'  ) }
         time   = { check_max( 16.h  * task.attempt, 'time'    ) }
     }
     withLabel:process_long {
         time   = { check_max( 20.h  * task.attempt, 'time'    ) }
     }
     withLabel:process_high_memory {
+        cpus   = { check_max( 8    * task.attempt, 'cpus'    ) }
         memory = { check_max( 200.GB * task.attempt, 'memory' ) }
+        time   = { check_max( 8.h  * task.attempt, 'time'    ) }
     }
     withLabel:error_ignore {
         errorStrategy = 'ignore'

conf/scflow_analysis.config

@@ -73,7 +73,7 @@ params {
     // Options: Dimensionality Reduction
     reddim_input_reduced_dim         = 'PCA,Liger' // *
     reddim_reduction_methods         = 'tSNE,UMAP,UMAP3D' // *
-    reddim_vars_to_regress_out       = 'nCount_RNA,pc_mito' // *
+    reddim_vars_to_regress_out       = 'total_counts,pc_mito' // *
     // umap
     reddim_umap_pca_dims             = 30
     reddim_umap_n_neighbors          = 35
@@ -115,6 +115,7 @@ params {
     // Options: Celltype Annotation
     cta_clusters_colname             = 'clusters'
     cta_cells_to_sample              = 10000
+    cta_annotation_level	     = 2
 
     // Options: Celltype Metrics Report
     cta_unique_id_var                = 'manifest'
@@ -137,8 +138,8 @@ params {
     dge_ref_class                    = 'Control'
     dge_confounding_vars             = 'cngeneson' // *
     dge_random_effects_var           = 'null'
-    dge_fc_threshold                 = 1.1
-    dge_pval_cutoff                  = 0.05
+    dge_logFC_threshold              = 0.25
+    dge_padj_cutoff                  = 0.05
     dge_n_label                      = 5
     dge_force_run                    = 'false'
     dge_max_cores                    = 'null'
@@ -153,7 +154,8 @@ params {
     dirich_celltype_var              = 'cluster_celltype'
     dirich_dependent_var             = 'diagnosis'
     dirich_ref_class                 = 'Control'
-    dirich_var_order                 = 'null' // *
+    dirich_var_order                 = 'Control,Case' // *
+    dirich_confounding_vars          = 'null' // *		
 
     // Options: Plots (Reduced Dim)
     plotreddim_reduction_methods     = 'UMAP_Liger' // *

conf/test.config

@@ -20,8 +20,8 @@ params {
 
   input = "https://raw.githubusercontent.com/nf-core/test-datasets/scflow/refs/SampleSheet.tsv"
   manifest = "https://raw.githubusercontent.com/nf-core/test-datasets/scflow/refs/Manifest.txt"
-  ensembl_mappings = "https://raw.githubusercontent.com/nfancy/test-datasets/scflow/assets/ensembl_mappings.tsv"
-  ctd_path = "https://s3-eu-west-1.amazonaws.com/pfigshare-u-files/28033407/ctd_v1.zip"
+  ensembl_mappings = "https://raw.githubusercontent.com/nf-core/test-datasets/scflow/assets/ensembl_mappings.tsv"
+  ctd_path = "https://raw.githubusercontent.com/combiz/scFlowData/dev-nf/assets/ctd.zip"
   reddim_genes_yml = "https://raw.githubusercontent.com/nf-core/test-datasets/scflow/refs/reddim_genes.yml"
 
   reddimplot_pointsize = 1

modules.json

@@ -1,8 +1,5 @@
 {
     "name": "nf-core/scflow",
     "homePage": "https://github.com/nf-core/scflow",
-    "repos": {
-        "nf-core/modules": {
-        }
-    }
+    "repos": {}
 }