Fix assert in process_sparse_gt (#95)

* Fix assert in process_sparse_gt * add filter_variants_multiple test * Fix path to test data
PMBio · May 22, 2024 · 6198999 · 6198999
1 parent aa5c87a
commit 6198999
Show file tree

Hide file tree

Showing 9 changed files with 25 additions and 5 deletions.
diff --git a/deeprvat/preprocessing/preprocess.py b/deeprvat/preprocessing/preprocess.py
@@ -255,6 +255,7 @@ def process_sparse_gt(
     if chromosomes is not None:
         chromosomes = [f"chr{chrom}" for chrom in chromosomes.split(",")]
         variants = variants[variants["chrom"].isin(chromosomes)]
+
     total_variants = len(variants)
     if len(exclude_variants) > 0:
         variant_exclusion_files = [
@@ -267,17 +268,15 @@ def process_sparse_gt(
             ],
             ignore_index=True,
         )
-        if chromosomes is not None:
-            variants_to_exclude = variants_to_exclude[
-                variants_to_exclude["chrom"].isin(chromosomes)
-            ]
+
         variants_to_exclude = variants_to_exclude.drop_duplicates(ignore_index=True)
         variant_ids_to_exclude = pd.merge(
             variants_to_exclude, variants, validate="1:1"
         )["id"]
+
         variants = variants[~variants["id"].isin(variant_ids_to_exclude)]
         if not skip_sanity_checks:
-            assert total_variants - len(variants) == len(variants_to_exclude)
+            assert total_variants - len(variants) == len(variant_ids_to_exclude)
 
     logging.info(f"Dropped {total_variants - len(variants)} variants")
     logging.info(f"...done ({time.time() - start_time} s)")

diff --git a/...rocessing/test_data/process_sparse_gt/filter_variants_multiple/expected/expected_data.npz b/...rocessing/test_data/process_sparse_gt/filter_variants_multiple/expected/expected_data.npz
diff --git a/...ocessing/test_data/process_sparse_gt/filter_variants_multiple/input/qc/input_c1_b1.tsv.gz b/...ocessing/test_data/process_sparse_gt/filter_variants_multiple/input/qc/input_c1_b1.tsv.gz
diff --git a/...ocessing/test_data/process_sparse_gt/filter_variants_multiple/input/qc/input_c1_b2.tsv.gz b/...ocessing/test_data/process_sparse_gt/filter_variants_multiple/input/qc/input_c1_b2.tsv.gz
diff --git a/.../preprocessing/test_data/process_sparse_gt/filter_variants_multiple/input/samples_chr.csv b/.../preprocessing/test_data/process_sparse_gt/filter_variants_multiple/input/samples_chr.csv
@@ -0,0 +1,11 @@
+100096
+100097
+100099
+100100
+100101
+100102
+100103
+100104
+100105
+100106
+100107
diff --git a/...t_data/process_sparse_gt/filter_variants_multiple/input/sparse_gt/chr1/input_c1_b1.tsv.gz b/...t_data/process_sparse_gt/filter_variants_multiple/input/sparse_gt/chr1/input_c1_b1.tsv.gz
diff --git a/...preprocessing/test_data/process_sparse_gt/filter_variants_multiple/input/variants.parquet b/...preprocessing/test_data/process_sparse_gt/filter_variants_multiple/input/variants.parquet
diff --git a/.../preprocessing/test_data/process_sparse_gt/filter_variants_multiple/input/variants.tsv.gz b/.../preprocessing/test_data/process_sparse_gt/filter_variants_multiple/input/variants.tsv.gz
diff --git a/tests/preprocessing/test_preprocess.py b/tests/preprocessing/test_preprocess.py
@@ -51,6 +51,16 @@ def load_h5_archive(h5_path):
             ],
             "genotypes_chr1.h5",
         ),
+        (
+            "filter_variants_multiple",
+            [
+                "--chromosomes",
+                "1",
+                "--exclude-variants",
+                f"{(tests_data_dir / 'process_sparse_gt/filter_variants_multiple/input/qc').as_posix()}",
+            ],
+            "genotypes_chr1.h5",
+        ),
         (
             "filter_samples_minimal",
             [
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,11 @@
		100096
		100097
		100099
		100100
		100101
		100102
		100103
		100104
		100105
		100106
		100107