Merge pull request #38 from Al-Murphy/main

Bug fix: handle cases where no genes found in species
neurogenomics · Dec 21, 2023 · 38ab713 · 38ab713
2 parents 4c643d5 + d5eb75c
commit 38ab713
Show file tree

Hide file tree

Showing 4 changed files with 63 additions and 43 deletions.
diff --git a/NEWS.md b/NEWS.md
@@ -5,6 +5,11 @@
 * Bump version for Bioc 3.19.
 * Use GHCR instead of Dockerhub, update vignette accordingly.
 
+## Bug fixes
+
+* Catch case where no genes are found in the species in `map_genes()` & test 
+  case added.
+
 # orthogene  1.7.3
 
 ## New features

diff --git a/R/map_genes.R b/R/map_genes.R
@@ -74,7 +74,9 @@ map_genes <- function(genes,
         )
     }
     #### Drop NAs ####
-    if (isTRUE(drop_na)) {
+    # !is.null(syn_map) catches where no hit genes are found in the species
+    # otherwise map_genes() will give an error rather than returning NULL
+    if (isTRUE(drop_na) && !is.null(syn_map)) {
         syn_map <- remove_all_nas(
             dat = syn_map,
             col_name = "name",

diff --git a/R/map_orthologs.R b/R/map_orthologs.R
@@ -56,48 +56,51 @@ map_orthologs <- function(genes,
         )
         genes <- syn_map$name
     }
-    #### Select mapping method ####
-    #### User-supplied mapping ####
-    if(!is.null(gene_map)){ 
-        gene_map <- map_orthologs_custom(gene_map = gene_map,
-                                         input_species = input_species,
-                                         output_species = output_species, 
-                                         input_col = input_col,
-                                         output_col = output_col, 
-                                         verbose = verbose)
-    }
-    # Both methods will return a dataframe with at least the columns
-    # "input_gene" and "ortholog_gene"
-    #### gprofiler ####
-    if (methods_opts(method = method, gprofiler_opts = TRUE)) {
-        gene_map <- map_orthologs_gprofiler(
-            genes = genes,
-            input_species = input_species,
-            output_species = output_species,
-            mthreshold = mthreshold,
-            verbose = verbose,
-            ...
-        )
-    }
-    #### homologene ####
-    if (methods_opts(method = method, homologene_opts = TRUE)) {
-        gene_map <- map_orthologs_homologene(
-            genes = genes,
-            input_species = input_species,
-            output_species = output_species,
-            verbose = verbose,
-            ...
-        )
-    }
-    #### babelgene ####
-    if (methods_opts(method = method, babelgene_opts = TRUE)) {
-        gene_map <- map_orthologs_babelgene(
-            genes = genes,
-            input_species = input_species,
-            output_species = output_species,
-            verbose = verbose,
-            ...
-        )
+    #deal with case where no genes found for species
+    if(!is.null(genes)){
+      #### Select mapping method ####
+      #### User-supplied mapping ####
+      if(!is.null(gene_map)){ 
+          gene_map <- map_orthologs_custom(gene_map = gene_map,
+                                           input_species = input_species,
+                                           output_species = output_species, 
+                                           input_col = input_col,
+                                           output_col = output_col, 
+                                           verbose = verbose)
+      }
+      # Both methods will return a dataframe with at least the columns
+      # "input_gene" and "ortholog_gene"
+      #### gprofiler ####
+      if (methods_opts(method = method, gprofiler_opts = TRUE)) {
+          gene_map <- map_orthologs_gprofiler(
+              genes = genes,
+              input_species = input_species,
+              output_species = output_species,
+              mthreshold = mthreshold,
+              verbose = verbose,
+              ...
+          )
+      }
+      #### homologene ####
+      if (methods_opts(method = method, homologene_opts = TRUE)) {
+          gene_map <- map_orthologs_homologene(
+              genes = genes,
+              input_species = input_species,
+              output_species = output_species,
+              verbose = verbose,
+              ...
+          )
+      }
+      #### babelgene ####
+      if (methods_opts(method = method, babelgene_opts = TRUE)) {
+          gene_map <- map_orthologs_babelgene(
+              genes = genes,
+              input_species = input_species,
+              output_species = output_species,
+              verbose = verbose,
+              ...
+          )
+      }
     }
     #### Check is already in the same species ####
     if(isFALSE(standardise_genes) &&

diff --git a/tests/testthat/test-map_genes.R b/tests/testthat/test-map_genes.R
@@ -58,4 +58,14 @@ test_that("map_genes works", {
     testthat::expect_gte(nrow(mapped_human), 3)
     testthat::expect_gte(nrow(mapped_mouse), total_genes)
     testthat::expect_gte(nrow(mapped_zebrafish), 3)
+
+    #### Test where genes aren't found in species ####
+    #should return NULL not fail
+    genes <- c("a",'list','of','fake','genes')
+    spec_hits <- map_genes(
+      genes = genes,
+      species = 'yeast',
+      drop_na = TRUE,
+      verbose = FALSE)$name
+    testthat::expect_null(spec_hits)
 })