fix read bgen

TARGENE · Jun 7, 2024 · 210f521 · 210f521
1 parent f0c00a3
commit 210f521
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 2 deletions.
diff --git a/src/utils.jl b/src/utils.jl
@@ -9,9 +9,14 @@ function keep_only_imputed(associations, bgen_prefix, chr)
     return innerjoin(associations, imputed_rsids, on=:SNP)
 end
 
+get_bgen_chromosome(bgen_files, chr) = only(filter(
+    x -> endswith(x, Regex(string("[^0-9]", chr, ".bgen"))), 
+    bgen_files
+    ))
+
 function read_bgen_chromosome(bgen_prefix, chr)
-    all_bgens = files_matching_prefix(bgen_prefix)
-    bgen_file = only(filter(x -> endswith(x, string(chr, ".bgen")), all_bgens))
+    bgen_files = files_matching_prefix(bgen_prefix)
+    bgen_file = get_bgen_chromosome(bgen_files, chr)
     return TargeneCore.read_bgen(bgen_file)
 end
 

diff --git a/test/utils.jl b/test/utils.jl
@@ -122,6 +122,20 @@ end
         1 0 1 0
         0 1 0 1
     ]
+    # Test get_bgen_chromosome
+    bgen_files = [
+        "ukb_53116_chr11.bgen",   
+        "ukb_53116_chr12.bgen",
+        "ukb_53116_chr12.bgen.bgi",              
+        "ukb_53116_chr12.sample",
+        "ukb_53116_chr2.bgen",
+        "ukb_53116_chr2.bgen.bgi",
+        "ukb_53116_chr2.sample"
+    ]
+    @test Simulations.get_bgen_chromosome(bgen_files, 2) == "ukb_53116_chr2.bgen"
+    @test Simulations.get_bgen_chromosome(bgen_files, 12) == "ukb_53116_chr12.bgen"
+    @test Simulations.get_bgen_chromosome(bgen_files, 11) == "ukb_53116_chr11.bgen"
+
     # Test read_bgen_chromosome(bgen_prefix, chr)
     bgen_prefix = joinpath(TARGENCORE_TESTDIR, "data", "ukbb", "imputed", "ukbb")
     b = Simulations.read_bgen_chromosome(bgen_prefix, 12)