fix: align karyotype implementation with config-driven patterns and repo conventions

Author: adilraza99

malariagen_data/anoph/karyotype.py

@@ -70,9 +70,12 @@ def _require_karyotype_analysis(self):
     def load_inversion_tags(self, inversion: inversion_param) -> pd.DataFrame:
         self._require_karyotype_analysis()
 
+        filename = self.config.get(
+            "KARYOTYPE_TAG_SNPS_FILENAME", "karyotype_tag_snps.csv"
+        )
         path = (
             f"{self._base_path}/{self._major_version_path}"
-            f"/karyotype/{self._karyotype_analysis}/karyotype_tag_snps.csv"
+            f"/snp_karyotype/{self._karyotype_analysis}/{filename}"
         )
         with self._fs.open(path) as f:
             df_tag_snps = pd.read_csv(f, sep=",")

tests/anoph/conftest.py

@@ -1154,7 +1154,10 @@ def init_config(self):
             "SITE_ANNOTATIONS_ZARR_PATH": "reference/genome/agamp4/Anopheles-gambiae-PEST_SEQANNOTATION_AgamP4.12.zarr",
             "DEFAULT_AIM_ANALYSIS": "20220528",
             "DEFAULT_SITE_FILTERS_ANALYSIS": "dt_20200416",
+            # Simulated placeholder; real value will be set in production config.
             "DEFAULT_KARYOTYPE_ANALYSIS": "20231213",
+            "KARYOTYPE_INVERSIONS": ["2La", "2Rb"],
+            "KARYOTYPE_TAG_SNPS_FILENAME": "karyotype_tag_snps.csv",
             "DEFAULT_COHORTS_ANALYSIS": "20230516",
             "SITE_MASK_IDS": ["gamb_colu_arab", "gamb_colu", "arab"],
             "PHASING_ANALYSIS_IDS": ["gamb_colu_arab", "gamb_colu", "arab"],
@@ -1526,10 +1529,13 @@ def init_snp_sites(self):
 
     def init_karyotype_tags(self):
         analysis = self.config["DEFAULT_KARYOTYPE_ANALYSIS"]
+        inversions = self.config["KARYOTYPE_INVERSIONS"]
+        filename = self.config["KARYOTYPE_TAG_SNPS_FILENAME"]
 
         # Generate tag SNP data using positions from simulated SNP sites.
         tags = []
-        for contig, inversion in [("2R", "2Rb"), ("2L", "2La")]:
+        for inversion in inversions:
+            contig = inversion[:2]
             snp_pos = self.snp_sites[contig]["variants"]["POS"][:]
             snp_alt = self.snp_sites[contig]["variants"]["ALT"][:]
             n_tags = min(20, len(snp_pos))
@@ -1546,9 +1552,7 @@ def init_karyotype_tags(self):
                 )
 
         df = pd.DataFrame(tags)
-        path = (
-            self.bucket_path / "v3" / "karyotype" / analysis / "karyotype_tag_snps.csv"
-        )
+        path = self.bucket_path / "v3" / "snp_karyotype" / analysis / filename
         path.parent.mkdir(parents=True, exist_ok=True)
         df.to_csv(path, index=False)