Merge pull request #138 from adaptyvbio/pdb_format

fix: add a check for to_pdb

proteinflow/data/__init__.py

@@ -575,6 +575,9 @@ def rename_chains(self, chain_dict):
             A dictionary mapping old chain IDs to new chain IDs
 
         """
+        for chain in self.get_chains():
+            if chain not in chain_dict:
+                chain_dict[chain] = chain
         self._rename_chains({k: k * 5 for k in self.get_chains()})
         self._rename_chains({k * 5: v for k, v in chain_dict.items()})
 
@@ -988,6 +991,10 @@ def to_pdb(
             Title of the PDB file (by default either the protein id or "Untitled")
 
         """
+        if any([x[0].upper() != x for x in self.get_chains()]):
+            raise ValueError(
+                "Chain IDs must be single uppercase letters, please rename with `rename_chains` before saving."
+            )
         pdb_builder = PDBBuilder(
             self,
             only_ca=only_ca,

proteinflow/data/utils.py

@@ -256,7 +256,8 @@ def _make_SEQRES(self):
     def save_pdb(self, path, title="UntitledProtein"):
         """Write out the generated PDB file as a string to the specified path."""
         with open(path, "w") as outfile:
-            outfile.write(self.get_pdb_string(title))
+            text = self.get_pdb_string(title)
+            outfile.write(text)
 
 
 def _split_every(n, iterable):