forbid too aggressive branch swapping in find_isomorphs

CHANGELOG.md

@@ -121,7 +121,7 @@
 ##### Added ✨
 - Natively support narrow linkage ambiguity in `categorical_node_match_wildcard`; that means you can use things like "Gal(b1-3/4)GlcNAc" with `subgraph_isomorphism` or `compare_glycans` (as well as all functions using these core functions) and it will only return True for "Gal(b1-3)GlcNAc", "Gal(b1-4)GlcNAc", and "Gal(b1-?)GlcNAc" (b94744e)
 - Added `build_wildcard_cache` for a central handling of wildcard mapping that can also be cached (a98461f)
-- `compare_glycans` now also has the `return_matches` keyword argument that allows for a retrieval of the node mapping if the glycans are isomorphic
+- `compare_glycans` now also has the `return_matches` keyword argument that allows for a retrieval of the node mapping if the glycans are isomorphic (7c510c9)
 
 ##### Changed 🔄
 - Ensured that `compare_glycans` is 100% order-specific, never matching something like ("Gal(b1-4)GlcNAc", "GlcNAc(b1-4)Gal") (5a99d6b)
@@ -148,7 +148,7 @@
 - `get_coordinates_and_labels` now internally uses `motif.processing.choose_correct_isoform` to reorder the glycan for drawing (41bb1a1)
 - Improved console drawing quality controlled by `display_svg_with_matplotlib` and image quality in Excel cells using `plot_glycans_excel` (a64f694)
 - `draw_chem2d` and `draw_chem3d` will now detect whether the user is in a Jupyter environment and, if not, plot to the Matplotlib console (c3a7f64)
-- `process_per_residue` now will re-order the `per_residue` list in the same way as the glycan is re-ordered for drawing with `GlycoDraw`
+- `process_per_residue` now will re-order the `per_residue` list in the same way as the glycan is re-ordered for drawing with `GlycoDraw` (7c510c9)
 
 ##### Deprecated ⚠️
 - Deprecated `hex_circumference`, the functionality is now available within `draw_hex` with the new keyword argument "outline_only" (4f1ccfa)

glycowork/glycan_data/loader.py

@@ -252,10 +252,10 @@ def download_model(file_id: str, # Google Drive file ID
   "Download the model weights file from Google Drive"
   file_id = file_id.split('/d/')[1].split('/view')[0]
   url = f'https://drive.google.com/uc?id={file_id}'
-  response = requests.get(url, stream=True, timeout=10)
+  response = requests.get(url, stream = True, timeout = 10)
   if response.status_code == 200:
     with open(local_path, 'wb') as f:
-      for chunk in response.iter_content(chunk_size=8192):
+      for chunk in response.iter_content(chunk_size = 8192):
         f.write(chunk)
     print("Download completed.")
   else:
@@ -360,8 +360,7 @@ def deserialize(cls, path: str # file path to load serialized data
 serializer = DataFrameSerializer()
 
 
-def count_nested_brackets(
-    s: str,
+def count_nested_brackets(s: str,
     length: bool = False
     ) -> int:
   count = 0

glycowork/motif/draw.py

@@ -1306,7 +1306,7 @@ def draw_chem3d(
   atom_colors = {k: ['#ECECEC'] if len(v) > 1 else v for k, v in atom_colors.items()}
 
   if pdb_file:
-    mol = MolFromPDBFile(pdb_file)
+    mol = MolFromPDBFile(str(pdb_file))
   else:
     mol = AddHs(mol)
     EmbedMolecule(mol)

glycowork/motif/processing.py

@@ -90,8 +90,8 @@ def get_possible_linkages(wildcard: str, # Pattern to match, ? can be wildcard
     numbers = re.search(r'-(\d+(?:/\d+)*)', wildcard).group(1).split('/')
     base_pattern = f"{prefix}-({('|'.join(numbers))}|\\?)"
     return {l for l in linkage_list if re.compile(f'^{base_pattern}$').fullmatch(l)} | \
-           ({f"{wildcard[:wildcard.index('-')]}-{'/'.join(sorted(combo))}" 
-            for combo in combinations(numbers, r = 2)} if len(numbers) > 2 else set())
+           ({f"{wildcard[:wildcard.index('-')]}-{'/'.join(sorted(combo))}"
+             for combo in combinations(numbers, r = 2)} if len(numbers) > 2 else set())
   pattern = f"^{wildcard.replace('?', '[ab1-9?]')}$"
   return {l for l in linkage_list if re.compile(pattern).fullmatch(l)}
 
@@ -128,7 +128,7 @@ def bracket_removal(glycan_part: str # Residual part of glycan from glycan_to_gr
 def find_isomorphs(glycan: str # Glycan in IUPAC-condensed format
                  ) -> List[str]: # List of isomorphic glycan notations
   "Returns a set of isomorphic glycans by swapping branches"
-  if '[' not in glycan or glycan.index('[') == 0:
+  if '[' not in glycan or ']' not in glycan or glycan.index('[') == 0:
     return [glycan]
   floaty = False
   if '{' in glycan:
@@ -162,18 +162,22 @@ def find_isomorphs(glycan: str # Glycan in IUPAC-condensed format
   temp = set()
   for k in out_list:
     if k.count('[') >= 3 and k.count('][') >= 2:
-      m = re.search(r'^(.*?)\[(.*?)\]\[(.*?)\]\[(.*?)\](.*?)$', k)
-      if m and not bool(re.search(r'\[[^\]]+\[', k)):
-        main, b1, b2, b3, rest = m.groups()
-        branches = [main, b1, b2, b3]
-        # Generate all 24 permutations
-        for p in permutations(range(4)):
-          # First element is main chain (no brackets), rest get brackets
-          result = branches[p[0]]
-          for idx in p[1:]:
-            result += f"[{branches[idx]}]"
-          result += rest
-          temp.add(result)
+      groups = re.finditer(r'((?:^[^[]+|^)\[[^[\]]*\]\[[^[\]]*\]\[[^[\]]*\](?=[A-Z]|$))', k)
+      for g in groups:
+        branch_section = g.group(1)
+        if not bool(re.search(r'\[[^\]]+\[', branch_section)):
+          m = re.search(r'^(.*?)\[(.*?)\]\[(.*?)\]\[(.*?)\]', branch_section)
+          if m:
+            main, b1, b2, b3 = m.groups()
+            branches = [main, b1, b2, b3]
+            # Generate all 24 permutations
+            for p in permutations(range(4)):
+              # First element is main chain (no brackets), rest get brackets
+              result = k[:g.start(1)] + branches[p[0]]
+              for idx in p[1:]:
+                result += f"[{branches[idx]}]"
+              result += k[g.end(1):]
+              temp.add(result)
   out_list.update(temp)
   temp = set()
   # Starting branch swapped with next side branch again to also include double branch swapped isomorphs
@@ -182,7 +186,7 @@ def find_isomorphs(glycan: str # Glycan in IUPAC-condensed format
     if k.count('[') > 1 and k.index('[') > 0 and find_nth(k, '[', 2) > k.index(']') and (find_nth(k, ']', 2) < find_nth(k, '[', 3) or k.count('[') == 2):
       temp.add(re.sub(r'^(.*?)\[(.*?)\](.*?)\[(.*?)\]', r'\4[\1[\2]\3]', k, 1))
   out_list.update(temp)
-  out_list = {k for k in out_list if not any([j in k for j in ['[[', ']]']])}
+  out_list = {k for k in out_list if not any([j in k for j in ['[[', ']]']]) and k.index(']') > k.index('[')}
   if floaty:
     out_list = {floaty+k for k in out_list}
   return list(out_list)

tests/test_core_functions.py

@@ -1127,6 +1127,7 @@ def test_choose_correct_isoform():
     result = choose_correct_isoform("Gal(a1-3)[Fuc(a1-2)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-6)]Gal(b1-4)GlcNAc(b1-6)[Gal(a1-3)[Fuc(a1-2)]Gal(b1-4)GlcNAc(b1-3)]GalNAc")
     assert result == "Fuc(a1-2)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-6)[Gal(a1-3)]Gal(b1-4)GlcNAc(b1-6)[Fuc(a1-2)[Gal(a1-3)]Gal(b1-4)GlcNAc(b1-3)]GalNAc"
     assert choose_correct_isoform("Xyl(b1-2)[Man(a1-3)][Man(a1-6)][GlcNAc(b1-4)]Man(b1-4)GlcNAc(b1-4)GlcNAc") == "Xyl(b1-2)[Man(a1-3)][GlcNAc(b1-4)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc"
+    assert choose_correct_isoform('Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)][Fuc(a1-6)]GlcNAc') == "Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)][Fuc(a1-6)]GlcNAc"
     # Mode for GlycoDraw
     result = choose_correct_isoform("Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)[Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-4)]Man(a1-3)[Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)[Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)]Man(a1-6)][GlcNAc(b1-4)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc", order_by="linkage")
     assert result == "Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)[Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-4)]Man(a1-3)[GlcNAc(b1-4)][Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)[Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc"