Include additional parameters

.github/workflows/pull_request.yml

@@ -1,6 +1,6 @@
 name: CI
 
-on: [push]
+on: [push, pull_request]
 
 jobs:
   build:

scmap-cli-post-install-tests.bats

@@ -110,6 +110,29 @@
     [ -f  "$closest_cells_similarities_text_file" ]
 }
 
+@test "For each cell in a query dataset, search for the nearest neighbours by cosine distance within a collection of reference datasets. Includes additional parameters for refining the results" {
+    if [ "$use_existing_outputs" = 'true' ] && [ -f "$closest_cells_similarities_text_file_2" ]; then
+        skip "$closest_cells_similarities_text_file_2 exists and use_existing_outputs is set to 'true'"
+    fi
+
+    run rm -rf $closest_cells_similarities_text_file_2 && scmap-scmap-cell.R\
+                                                    -i $index_cell_sce\
+                                                    -p $test_sce_processed\
+                                                    --number-nearest-neighbours $cell_number_nearest_neighbours\
+                                                    --nearest-neighbours-threshold $cell_nearest_neighbours_threshold\
+                                                    --threshold $cell_similarity_threshold\
+                                                    --cluster-col $cluster_col\
+                                                    --output-object-file $closest_cells_clusters_sce\
+                                                    --output-clusters-text-file $closest_cells_clusters_tsv\
+                                                    --closest-cells-text-file $closest_cells_text_file\
+                                                    --closest-cells-similarities-text-file $closest_cells_similarities_text_file_2
+
+    echo "status = ${status}"
+    echo "output = ${output}"
+
+    [ "$status" -eq 0 ]
+    [ -f  "$closest_cells_similarities_text_file_2" ]
+}
 
 @test "Obtain standard output" {
     if [ "$use_existing_outputs" = 'true' ] && [ -f "$scmap_output_tbl" ]; then

scmap-cli-post-install-tests.sh

@@ -66,6 +66,7 @@ export index_cell_sce=$output_dir'/index_cell.rds'
 export project_cell_sce=$output_dir'/project_cell.rds'
 export closest_cells_text_file=$output_dir'/closest_cells.tsv'
 export closest_cells_similarities_text_file=$output_dir'/closest_cells_similarities.tsv'
+export closest_cells_similarities_text_file_2=$output_dir'/closest_cells_similarities_2.tsv'
 export closest_cells_clusters_sce=$output_dir'/closest_cells_clusters.rds'
 export closest_cells_clusters_tsv=$output_dir'/closest_cells_clusters.tsv'
 export scmap_output_tbl=$output_dir'/scmap_output_tbl.txt'
@@ -84,6 +85,8 @@ export random_seed=1
 export cells_number_chunks=50
 export cells_number_clusters=9
 export cell_number_nearest_neighbours=5
+export cell_nearest_neighbours_threshold=1
+export cell_similarity_threshold=0.4
 
 ################################################################################
 # Test individual scripts
@@ -99,4 +102,4 @@ export use_existing_outputs
 tests_file="${script_name%.*}".bats
 
 # Execute the bats tests
-$tests_file
+$tests_file

scmap-scmap-cell.R

@@ -31,6 +31,20 @@ option_list = list(
     type = 'numeric',
     help = 'A positive integer specifying the number of nearest neighbours to find.'
   ),
+  make_option(
+    c("-n", "--nearest-neighbours-threshold"),
+    action = "store",
+    default = 3,
+    type = 'numeric',
+    help = 'A positive integer specifying the number of matching nearest neighbours required to label a cell.'
+  ),
+  make_option(
+    c("-r", "--threshold"),
+    action = "store",
+    default = 0.7,
+    type = 'numeric',
+    help = 'Threshold on similarity (or probability for SVM and RF).'
+  ),
   make_option(
     c("-c", "--cluster-col"),
     action = "store",
@@ -104,7 +118,9 @@ if (opt$cluster_col %in% colnames(colData(index_sce))){
     scmapCell_results, 
     list(
       as.character(colData(index_sce)[[opt$cluster_col]])
-    )
+    ),
+    w=opt$nearest_neighbours_threshold,
+    threshold=opt$threshold
   )
 
   # Output format anticipates multiple input indexes, let's assume a single input

scmap-scmap-cluster.R

@@ -73,7 +73,8 @@ scmapCluster_results <- scmapCluster(
   projection = project_sce, 
   index_list = list(
     metadata(index_sce)$scmap_cluster_index
-  )
+  ),
+  threshold = opt$threshold
 )
 
 # Output format anticipates multiple input indexes, let's assume a single input

scmap-select-features.R

@@ -55,10 +55,10 @@ if ( ! file.exists(opt$input_object_file)){
 SingleCellExperiment <- readRDS(opt$input_object_file)
 
 if (is.na(opt$output_plot_file)){
-  SingleCellExperiment <- selectFeatures(SingleCellExperiment, suppress_plot = TRUE)
+  SingleCellExperiment <- selectFeatures(SingleCellExperiment, n_features = opt$n_features, suppress_plot = TRUE)
 }else{
   png(file = opt$output_plot_file)
-  SingleCellExperiment <- selectFeatures(SingleCellExperiment, suppress_plot = FALSE)
+  SingleCellExperiment <- selectFeatures(SingleCellExperiment, n_features = opt$n_features, suppress_plot = FALSE)
   dev.off()
 }