prepare for hydroxyl-less monomers

Author: lucidrains

alphafold3_pytorch/inputs.py

@@ -471,21 +471,9 @@ def map_int_or_string_indices_to_mol(
     else:
         entries = [entries[s] for s in indices]
 
-    # for all peptides or nucleotide except last, remove hydroxl
+    # gather Chem.Mol(s)
 
-    mols = []
-
-    for idx, entry in enumerate(entries):
-        is_last = idx == (len(entries) - 1)
-
-        mol = entry[mol_keyname]
-
-        if chain and not is_last:
-            # hydroxyl oxygen to be removed should be the last atom
-            hydroxyl_idx = mol.GetNumAtoms() - 1
-            mol = remove_atom_from_mol(mol, hydroxyl_idx)
-
-        mols.append(mol)
+    mols = [entry[mol_keyname] for entry in entries]
 
     if not return_entries:
         return mols

alphafold3_pytorch/life.py

@@ -26,7 +26,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)C',
         first_atom_idx = 5,
         last_atom_idx = 2,
-        hydroxyl_idx = 4,
         distogram_atom_idx = 1,
         token_center_atom_idx = 1,
     ),
@@ -35,7 +34,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CCCNC(N)=N',
         first_atom_idx = 6,
         last_atom_idx = 3,
-        hydroxyl_idx = 5,
         distogram_atom_idx = 2,
         token_center_atom_idx = 2,
     ),
@@ -44,7 +42,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC(=O)N',
         first_atom_idx = 5,
         last_atom_idx = 2,
-        hydroxyl_idx = 4,
         distogram_atom_idx = 1,
         token_center_atom_idx = 1,
     ),
@@ -53,7 +50,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC(=O)O',
         first_atom_idx = 5,
         last_atom_idx = 2,
-        hydroxyl_idx = 8,
         distogram_atom_idx = 1,
         token_center_atom_idx = 1,
     ),
@@ -62,7 +58,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CS',
         first_atom_idx = 5,
         last_atom_idx = 2,
-        hydroxyl_idx = 4,
         distogram_atom_idx = 1,
         token_center_atom_idx = 1,
     ),
@@ -71,7 +66,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CCC(=O)N',
         first_atom_idx = 9,
         last_atom_idx = 6,
-        hydroxyl_idx = 8,
         distogram_atom_idx = 5,
         token_center_atom_idx = 5,
     ),
@@ -80,7 +74,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CCC(=O)O',
         first_atom_idx = 9,
         last_atom_idx = 6,
-        hydroxyl_idx = 8,
         distogram_atom_idx = 5,
         token_center_atom_idx = 5,
     ),
@@ -89,7 +82,6 @@ def is_unique(arr):
         # template_smile = 'NCC=O',
         first_atom_idx = 4,
         last_atom_idx = 1,
-        hydroxyl_idx = 3,
         distogram_atom_idx = 0,
         token_center_atom_idx = 0,
     ),
@@ -98,7 +90,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC1=CNC=N1',
         first_atom_idx = 10,
         last_atom_idx = 7,
-        hydroxyl_idx = 9,
         distogram_atom_idx = 0,
         token_center_atom_idx = 0,
     ),
@@ -107,7 +98,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)C(CC)C',
         first_atom_idx = 8,
         last_atom_idx = 5,
-        hydroxyl_idx = 7,
         distogram_atom_idx = 0,
         token_center_atom_idx = 0,
     ),
@@ -116,7 +106,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC(C)C',
         first_atom_idx = 8,
         last_atom_idx = 5,
-        hydroxyl_idx = 7,
         distogram_atom_idx = 4,
         token_center_atom_idx = 4,
     ),
@@ -125,7 +114,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CCCCN',
         first_atom_idx = 9,
         last_atom_idx = 6,
-        hydroxyl_idx = 8,
         distogram_atom_idx = 5,
         token_center_atom_idx = 5,
     ),
@@ -134,7 +122,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CCSC',
         first_atom_idx = 8,
         last_atom_idx = 5,
-        hydroxyl_idx = 7,
         distogram_atom_idx = 4,
         token_center_atom_idx = 4,
     ),
@@ -143,7 +130,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC1=CC=CC=C1',
         first_atom_idx = 11,
         last_atom_idx = 8,
-        hydroxyl_idx = 10,
         distogram_atom_idx = 7,
         token_center_atom_idx = 7,
     ),
@@ -152,7 +138,6 @@ def is_unique(arr):
         # template_smile = 'N1C(C=O)CCC1',
         first_atom_idx = 3,
         last_atom_idx = 5,
-        hydroxyl_idx = 7,
         distogram_atom_idx = 2,
         token_center_atom_idx = 2,
     ),
@@ -161,7 +146,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CO',
         first_atom_idx = 5,
         last_atom_idx = 2,
-        hydroxyl_idx = 4,
         distogram_atom_idx = 1,
         token_center_atom_idx = 1,
     ),
@@ -170,7 +154,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)C(O)C',
         first_atom_idx = 6,
         last_atom_idx = 3,
-        hydroxyl_idx = 5,
         distogram_atom_idx = 2,
         token_center_atom_idx = 2,
     ),
@@ -179,7 +162,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC1=CNC2=C1C=CC=C2',
         first_atom_idx = 14,
         last_atom_idx = 11,
-        hydroxyl_idx = 13,
         distogram_atom_idx = 10,
         token_center_atom_idx = 10,
     ),
@@ -188,7 +170,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)CC1=CC=C(O)C=C1',
         first_atom_idx = 11,
         last_atom_idx = 8,
-        hydroxyl_idx = 10,
         distogram_atom_idx = 7,
         token_center_atom_idx = 7,
     ),
@@ -197,7 +178,6 @@ def is_unique(arr):
         # template_smile = 'NC(C=O)C(C)C',
         first_atom_idx = 7,
         last_atom_idx = 4,
-        hydroxyl_idx = 6,
         distogram_atom_idx = 3,
         token_center_atom_idx = 3,
     )
@@ -213,7 +193,6 @@ def is_unique(arr):
         complement = 'T',
         first_atom_idx = 20,
         last_atom_idx = 1,
-        hydroxyl_idx = 21,
         distogram_atom_idx = 4,
         token_center_atom_idx = 4,
     ),
@@ -223,7 +202,6 @@ def is_unique(arr):
         complement = 'G',
         first_atom_idx = 17,
         last_atom_idx = 1,
-        hydroxyl_idx = 19,
         distogram_atom_idx = 4,
         token_center_atom_idx = 4,
     ),
@@ -233,7 +211,6 @@ def is_unique(arr):
         complement = 'C',
         first_atom_idx = 21,
         last_atom_idx = 1,
-        hydroxyl_idx = 22,
         distogram_atom_idx = 4,
         token_center_atom_idx = 4,
     ),
@@ -243,7 +220,6 @@ def is_unique(arr):
         complement = 'A',
         first_atom_idx = 19,
         last_atom_idx = 11,
-        hydroxyl_idx = 20,
         distogram_atom_idx = 9,
         token_center_atom_idx = 8,
     )
@@ -256,7 +232,6 @@ def is_unique(arr):
         complement = 'U',
         first_atom_idx = 19,
         last_atom_idx = 11,
-        hydroxyl_idx = 20,
         distogram_atom_idx = 9,
         token_center_atom_idx = 9,
     ),
@@ -266,7 +241,6 @@ def is_unique(arr):
         complement = 'G',
         first_atom_idx = 17,
         last_atom_idx = 10,
-        hydroxyl_idx = 19,
         distogram_atom_idx = 8,
         token_center_atom_idx = 8,
     ),
@@ -276,7 +250,6 @@ def is_unique(arr):
         complement = 'C',
         first_atom_idx = 14,
         last_atom_idx = 7,
-        hydroxyl_idx = 16,
         distogram_atom_idx = 5,
         token_center_atom_idx = 5,
     ),
@@ -286,7 +259,6 @@ def is_unique(arr):
         complement = 'A',
         first_atom_idx = 18,
         last_atom_idx = 10,
-        hydroxyl_idx = 19,
         distogram_atom_idx = 8,
         token_center_atom_idx = 8,
     )
@@ -549,7 +521,6 @@ def mol_from_template_mmcif_file(
 
     atom_order[entry['first_atom_idx']] = -1
     atom_order[entry['last_atom_idx']] = 1e4
-    atom_order[entry['hydroxyl_idx']] = 1e4 + 1
 
     atom_reorder = atom_order.argsort()
 

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "alphafold3-pytorch"
-version = "0.1.118"
+version = "0.1.120"
 description = "Alphafold 3 - Pytorch"
 authors = [
     { name = "Phil Wang", email = "[email protected]" }