update cycombine

Author: ghar1821

src/methods/cycombine_no_controls_to_goal/config.vsh.yaml

@@ -48,9 +48,8 @@ resources:
   # The script of your component (required)
   - type: r_script
     path: script.R
-  # Additional resources your script needs (optional)
-  # - type: file
-  #   path: weights.pt
+  - type: r_script
+    path: /src/utils/helper_functions.R
 
 engines:
   - type: docker

src/methods/cycombine_no_controls_to_goal/script.R

@@ -3,16 +3,22 @@ requireNamespace("cyCombine", quietly = TRUE)
 
 ## VIASH START
 par <- list(
-    input = "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/unintegrated_censored.h5ad",
-    output = "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/output.h5ad",
-    som_grid_size = 8,
+    input = "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/censored_split1.h5ad",
+    output = "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/output_split1.h5ad",
+    som_grid_size = 4,
     rlen = 10
 )
-meta <- list(name = "cycombine_no_controls")
+meta <- list(
+    name = "cycombine_no_controls_to_goal",
+    resources_dir = "src/utils"
+)
 ## VIASH END
 
+source(paste0(meta$resources_dir, "/helper_functions.R"))
+
 cat("Reading input files\n")
-input_adata <- anndata::read_h5ad(par[["input"]])
+input_adata <- anndata::read_h5ad(par[["input"]]) |>
+  subset_nocontrols()
 
 cat("Preparing input Anndata and df\n")
 

src/methods/cycombine_one_control_to_goal/config.vsh.yaml

@@ -58,9 +58,8 @@ resources:
   # The script of your component (required)
   - type: r_script
     path: script.R
-  # Additional resources your script needs (optional)
-  # - type: file
-  #   path: weights.pt
+  - type: r_script
+    path: /src/utils/helper_functions.R
 
 engines:
   - type: docker

src/methods/cycombine_one_control_to_goal/script.R

@@ -3,14 +3,22 @@ requireNamespace("anndata", quietly = TRUE)
 
 ## VIASH START
 par <- list(
-    input = "resources_test/task_cyto_batch_integration/leomazzi_cyto_spleen_subset/unintegrated_censored.h5ad",
-    output = "resources_test/task_cyto_batch_integration/leomazzi_cyto_spleen_subset/output.h5ad"
+    input = "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/censored_split1.h5ad",
+    output = "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/output_split1.h5ad",
+    som_grid_size = 4,
+    rlen = 10
+)
+meta <- list(
+    name = "cycombine_one_control_to_goal",
+    resources_dir = "src/utils"
 )
-meta <- list(name = "cycombine_one_control")
 ## VIASH END
 
+source(paste0(meta$resources_dir, "/helper_functions.R"))
+
 cat("Reading input files\n")
-input_adata <- anndata::read_h5ad(par[["input"]])
+input_adata <- anndata::read_h5ad(par[["input"]]) |>
+    subset_onecontrol()
 
 cat("Preparing input Anndata and df\n")
 

src/utils/helper_functions.R

@@ -89,6 +89,32 @@ subset_nocontrols <- function(adata) {
     adata[adata$obs$is_control == 0, ]
 }
 
+
+#' Subset AnnData Object to Single Control
+#'
+#' @description
+#' Extracts no control samples and just one control sample from an AnnData object based on the specified index.
+#'
+#' @param adata An AnnData object containing multiple samples with control groups.
+#' @param which_control Integer specifying which control sample to extract (default: 1).
+#'
+#' @return
+#' An AnnData object subset to contain only the specified control sample.
+#'
+#' @examples
+#' \dontrun{
+#' control_sample <- subset_onecontrol(adata, which_control = 1)
+#' }
+#'
+subset_onecontrol <- function(adata, which_control = 1) {
+    if (!"is_control" %in% colnames(adata$obs)) {
+        stop("The column 'is_control' is not present in the adata object.")
+    }
+
+    # Subset the adata to keep cells where is_control == which_control
+    adata[adata$obs$is_control %in% c(which_control, 0), ]
+}
+
 #' Subsets the anndata object to only include markers that need to be
 #' (or have been) corrected.
 #' These markers are all the entries in adata.var where to_correct == TRUE.