Actually fix the thing

workflow/scripts/weighted_distances.py

@@ -41,7 +41,7 @@ def build_ancestor_variant_table(ancestor: Seq, reference: Seq, reference_name:
     pos = []
     alt = []
     for i in range(1, len(ancestor) + 1):
-        if i not in masked_positions:
+        if i not in masked_positions and ancestor[i-1] != reference[i-1]:
             pos.append(i)
             alt.append(reference[i-1])
     df = pd.DataFrame({"POS": pos, "ALT": alt})
@@ -78,9 +78,9 @@ def calc_fst_weir_cockerham(hs: float, ht: float) -> float:
     return (ht - hs) / ht if ht != 0 else 0
 
 
-def build_cache(variant_table: pd.DataFrame, reference: Seq):
+def build_cache(variant_table: pd.DataFrame, samples: List[str], reference: Seq):
     cache = {"freq": {}, "hz": {}}
-    for sample_name in variant_table["REGION"].unique():
+    for sample_name in set(samples):
         for position in variant_table["POS"].astype("Int64").unique():
             if sample_name not in cache["freq"]:
                 cache["freq"][sample_name] = {}
@@ -120,7 +120,7 @@ def calculate_sample_distances(positions: List[int], sample_name: str, samples:
 
 def calculate_distance_matrix(variant_table: pd.DataFrame, samples: List[str], reference: Seq) -> pd.DataFrame:
     positions = variant_table["POS"].astype("Int64").unique().tolist()
-    cache = build_cache(variant_table, reference)
+    cache = build_cache(variant_table, samples, reference)
     distance_matrix = {}
     for sample_name in samples:
         distance_matrix[sample_name] = calculate_sample_distances(positions, sample_name, samples, cache)