add test with variants that fail harmonisation

pgscatalog.core/tests/data/bad_harmonised.txt

@@ -0,0 +1,23 @@
+###PGS CATALOG SCORING FILE - see https://www.pgscatalog.org/downloads/#dl_ftp_scoring for additional information
+#format_version=2.0
+##POLYGENIC SCORE (PGS) INFORMATION
+#pgs_id=PGS003581
+#pgs_name=MTAG.Envs.LifetimeMDD
+#trait_reported=Major Depressive Disorder (Lifetime)
+#trait_mapped=major depressive disorder
+#trait_efo=MONDO_0002009
+#genome_build=GRCh37
+#variants_number=4861398
+#weight_type=beta
+##SOURCE INFORMATION
+#pgp_id=PGP000461
+#citation=Dahl A et al. bioRxiv (2022). doi:10.1101/2022.08.15.503980
+##HARMONIZATION DETAILS
+#HmPOS_build=GRCh38
+#HmPOS_date=2023-04-12
+#HmPOS_match_chr={""True"": null, ""False"": null}
+#HmPOS_match_pos={""True"": null, ""False"": null}
+rsID	effect_allele	other_allele	effect_weight	allelefrequency_effect	variant_description	hm_source	hm_rsID	hm_chr	hm_pos	hm_inferOtherAllele
+.	T	C	0.004040743	0.329683	VARIANT_ID=1:2556125_C_T	Unknown				
+.	T	C	0.004025092	0.329689	VARIANT_ID=1:2556548_C_T	Unknown				
+.	A	G	0.004013246	0.329686	VARIANT_ID=1:2556709_G_A	Unknown				

pgscatalog.core/tests/test_combine_cli.py

@@ -83,6 +83,30 @@ def lift_scorefiles(request):
     )
 
 
+@pytest.fixture(scope="package")
+def fail_harmonised(request):
+    """This scoring file only contains variants that failed harmonisation."""
+    return request.path.parent / "data" / "bad_harmonised.txt"
+
+
+def test_fail_harmonised(tmp_path, fail_harmonised):
+    """Variants that have failed harmonisation will be missing mandatory fields, but we should accept them as a special case and write out like normal"""
+    out_path = tmp_path / "combined.txt"
+    path = [str(fail_harmonised)]
+
+    args = [("pgscatalog-combine", "-s"), path, ("-o", str(out_path), "-t", "GRCh38")]
+    flargs = list(itertools.chain(*args))
+
+    with patch("sys.argv", flargs):
+        run()
+
+    # https://github.com/PGScatalog/pygscatalog/issues/55
+    assert out_path.exists()
+    with open(out_path, mode="rt") as f:
+        x = list(csv.reader(f, delimiter="\t"))
+        assert len(x) == 4  # 3 variants + header
+
+
 def test_combine_nonadditive(tmp_path, non_additive_scorefile_grch38):
     """Test normalising a single non-additive scoring file fails."""
     out_path = tmp_path / "combined.txt.gz"