Add support for writing CIF files, update tests and documentation

- Added support for writing mmCIF files
- Added CIF file support in stack and align modules
- Moved align and tmalign tests to tests/ directory
- Updated tests for Windows-Linux compatibility
- Updated AlphaFold database version to v6
- Removed non-atom lines from AF file
- Updated changelog and fixed documentation links

Author: juliav

biopandas/align/align.py

@@ -49,6 +49,8 @@ def filter_and_validate_chain(self, pdb, chain_id):
                 chain_col = 'label_asym_id'
             else:
                 raise ValueError("No recognized chain identifier column found in the ATOM dataframe.")
+        else:
+            chain_col = 'chain_id'
 
         filtered_pdb = deepcopy(pdb)
         filtered_atoms = pdb.df['ATOM'][pdb.df['ATOM'][chain_col].isin([chain_id])]

biopandas/align/tmalign.py

@@ -35,6 +35,11 @@ def __init__(self, tmalign_path: str=None):
             self.tmalign_path = os.path.join(path_script, './USalign')
         elif os.path.exists(os.path.join(path_script, './USalign.exe')):
             self.tmalign_path = os.path.join(path_script, './USalign.exe')
+        # if our script has 'tests' in it, the path changes to "../../biopandas/align/"
+        elif os.path.exists(os.path.join(path_script, '../../biopandas/align/USalign')):
+            self.tmalign_path = os.path.join(path_script, '../../biopandas/align/USalign')
+        elif os.path.exists(os.path.join(path_script, '../../biopandas/align/USalign.exe')):
+            self.tmalign_path = os.path.join(path_script, '../../biopandas/align/USalign.exe')
         else:
             raise ValueError("Please provide the path to the TMalign executable.")
 

biopandas/mmcif/engines.py

@@ -83,7 +83,6 @@
     "occupancy": "occupancy",
     "B_iso_or_equiv": "b_factor",
     "type_symbol": "element_symbol",
-    "label_asym_id": "chain_id"
 }
 
 MMCIF_PDB_NONEFIELDS: List[str] = [

biopandas/mmcif/pandas_mmcif.py

@@ -166,8 +166,9 @@ def fetch_mmcif(
             uniprot_id is not None and source == "pdb"
         )
         invalid_input_combination_2 = pdb_code is not None and source in {
-            "alphafold2-v3",
-            "alphafold2-v4",
+            "alphafold2-v6",
+            # "alphafold2-v3", # deprecated
+            # "alphafold2-v4", # deprecated
         }
 
         if invalid_input_identifier_1 or invalid_input_identifier_2:
@@ -196,6 +197,11 @@ def fetch_mmcif(
             self.mmcif_path, self.mmcif_text = self._fetch_af2(
                 uniprot_id, af2_version
             )
+        elif source == "alphafold2-v6":
+            af2_version = 6
+            self.mmcif_path, self.mmcif_text = self._fetch_af2(
+                uniprot_id, af2_version
+            )
         else:
             raise ValueError(
                 f"Invalid source: {source}."

biopandas/pdb/pandas_pdb.py

@@ -138,7 +138,8 @@ def fetch_pdb(
 
                 source : str
                     The source to retrieve the structure from
-                    (`"pdb"`, `"alphafold2-v3"`, `"alphafold2-v4"`(latest)).
+                    # (`"pdb"`, `"alphafold2-v3"`, `"alphafold2-v4"`(latest)). #deprecated
+                     (`"pdb"`, `"alphafold2-v6"`(latest)).
                     Defaults to `"pdb"`.
 
                 Returns
@@ -151,8 +152,9 @@ def fetch_pdb(
         invalid_input_identifier_2 = pdb_code is not None and uniprot_id is not None
         invalid_input_combination_1 = uniprot_id is not None and source == "pdb"
         invalid_input_combination_2 = pdb_code is not None and source in {
-            "alphafold2-v3",
-            "alphafold2-v4",
+            "alphafold2-v6",
+            # "alphafold2-v3",
+            # "alphafold2-v4",
         }
 
         if invalid_input_identifier_1 or invalid_input_identifier_2:
@@ -173,6 +175,9 @@ def fetch_pdb(
         elif source == "alphafold2-v4":
             af2_version = 4
             self.pdb_path, self.pdb_text = self._fetch_af2(uniprot_id, af2_version)
+        elif source == "alphafold2-v6":
+            af2_version = 6
+            self.pdb_path, self.pdb_text = self._fetch_af2(uniprot_id, af2_version)
         elif source == "pdb":
             self.pdb_path, self.pdb_text = self._fetch_pdb(pdb_code)
         else:
@@ -454,6 +459,7 @@ def _fetch_af2(uniprot_id: str, af2_version: int = 3):
         """Load PDB file from https://alphafold.ebi.ac.uk/."""
         txt = None
         url = f"https://alphafold.ebi.ac.uk/files/AF-{uniprot_id.upper()}-F1-model_v{af2_version}.pdb"
+        print(url)
         try:
             response = urlopen(url)
             txt = response.read()

biopandas/stack/stack.py

@@ -118,7 +118,7 @@ def fetch_pdb(self, key: str=None, pdb_id: str=None, uniprot_id: str=None):
             if key is None:
                 key = pdb_id
         elif uniprot_id:
-            pdb.fetch_pdb(uniprot_id=uniprot_id, source="alphafold2-v4")
+            pdb.fetch_pdb(uniprot_id=uniprot_id, source="alphafold2-v6")
             if key is None:
                 key = uniprot_id
         else: