diff --git a/src/pg2_benchmark/dummy_data.py b/src/pg2_benchmark/dummy_data.py
index edd0e4a0..eb7ba4ad 100644
--- a/src/pg2_benchmark/dummy_data.py
+++ b/src/pg2_benchmark/dummy_data.py
@@ -110,9 +110,11 @@ def charge_mutations(sequence: str, n: int) -> str:
 
 def charge_ladder_dataset(n_rows: int = 200, seq_len: int = 20) -> pd.DataFrame:
     parent = _generate_sequence(seq_len)
-    sequences = [
+    # Deduplicate sequences
+    sequences = {
         charge_mutations(parent, random.randint(0, seq_len)) for _ in range(n_rows)
-    ]
+    }
+
     charge = [peptide_charge(seq) for seq in sequences]
     return pd.DataFrame({"sequence": sequences, "charge": charge})