should be fixed

Author: lucidrains

alphafold3_pytorch/alphafold3.py

@@ -4476,15 +4476,6 @@ def compute_weighted_lddt(
             else:
                 raise Exception(f"Invalid chain list {chains}")
 
-            if (atom_asym_id[b] == -1).all():
-                # TODO: Remove this check after fixing the batching bug in `batch_repeat_interleave()`
-                # (see https://github.com/lucidrains/alphafold3-pytorch/issues/158)
-                logger.warning(
-                    f"Found erroneous `atom_asym_id` element at index {b}. Returning null lDDT for this batch element."
-                )
-                weighted_lddt[b] = torch.tensor(1e-6, device=device)
-                continue
-
             type_chain_a = get_cid_molecule_type(
                 asym_id_a, atom_asym_id[b], atom_is_molecule_types[b], return_one_hot=False
             )