Merge pull request jeromekelleher#281 from jeromekelleher/recombinant…

…-update Construct a recombinant frankentype and test
szhan · Sep 19, 2024 · cdf35a9 · cdf35a9
2 parents 23f6b30 + 511b90d
commit cdf35a9
Show file tree

Hide file tree

Showing 8 changed files with 1,170 additions and 614 deletions.
diff --git a/notebooks/test_ts.ipynb b/notebooks/test_ts.ipynb
diff --git a/sc2ts/inference.py b/sc2ts/inference.py
diff --git a/sc2ts/info.py b/sc2ts/info.py
@@ -263,7 +263,9 @@ def node_counts(self):
 
     def _preprocess_nodes(self, show_progress):
         ts = self.ts
-        self.nodes_max_descendant_samples = max_descendant_samples(ts)
+        self.nodes_max_descendant_samples = max_descendant_samples(
+            ts, show_progress=show_progress
+        )
         self.nodes_date = np.zeros(ts.num_nodes, dtype="datetime64[D]")
         self.nodes_num_masked_sites = np.zeros(ts.num_nodes, dtype=np.int32)
         self.nodes_metadata = {}
@@ -367,7 +369,9 @@ def _preprocess_mutations(self, show_progress):
                     transversions.add((b1, b2))
 
         tree = ts.first()
-        iterator = tqdm.tqdm(np.arange(N), desc="Classifying mutations")
+        iterator = tqdm.tqdm(
+            np.arange(N), desc="Classifying mutations", disable=not show_progress
+        )
         for mut_id in iterator:
             tree.seek(self.mutations_position[mut_id])
             mutation_node = ts.mutations_node[mut_id]
@@ -657,7 +661,6 @@ def recombinants_summary(self):
         df["breakpoint_interval_right"] = interval_right
         df["max_descendant_samples"] = self.nodes_max_descendant_samples[df.node]
 
-
         df = df.set_index("node")
         # The MRCA table will contain duplicate rows here for two-or-more matches.
         # We just hacking for now to get things working well for two-parent recombs

diff --git a/sc2ts/metadata.py b/sc2ts/metadata.py
@@ -25,8 +25,8 @@ def __init__(self, path):
         logger.debug(f"Opened MetadataDb at {path} mode=ro")
 
     @staticmethod
-    def import_csv(csv_path, db_path):
-        df = pd.read_csv(csv_path, sep="\t")
+    def import_csv(csv_path, db_path, sep="\t"):
+        df = pd.read_csv(csv_path, sep=sep)
         db_path = pathlib.Path(db_path)
         if db_path.exists():
             db_path.unlink()

diff --git a/tests/conftest.py b/tests/conftest.py
@@ -72,7 +72,9 @@ def fx_ts_map(tmp_path, fx_data_cache, fx_metadata_db, fx_alignment_store):
     ]
     cache_path = fx_data_cache / f"{dates[-1]}.ts"
     if not cache_path.exists():
-        last_ts = sc2ts.initial_ts()
+        # These sites are masked out in all alignments in the initial data
+        # anyway; https://github.com/jeromekelleher/sc2ts/issues/282
+        last_ts = sc2ts.initial_ts([56, 57, 58, 59, 60])
         match_db = sc2ts.MatchDb.initialise(tmp_path / "match.db")
         for date in dates:
             last_ts = sc2ts.extend(