boltz-vsynthes 1.0.38__py3-none-any.whl → 1.0.40__py3-none-any.whl

boltz/data/parse/schema.py

@@ -1326,6 +1326,60 @@ def parse_boltz_schema(  # noqa: C901, PLR0915, PLR0912
                 "cyclic", False
             ), "Cyclic flag is not supported for ligands"
 
+        elif (entity_type == "ligand") and ("sdf" in items[0][entity_type]):
+            # Handle SDF file
+            sdf_path = Path(items[0][entity_type]["sdf"])
+            from boltz.data.parse.sdf import parse_sdf
+            target = parse_sdf(sdf_path, ccd, mol_dir)
+            mol = target["sequences"][0]["ligand"]["smiles"]
+
+            if affinity:
+                mol = standardize(mol)
+
+            mol = AllChem.MolFromSmiles(mol)
+            mol = AllChem.AddHs(mol)
+
+            # Set atom names
+            canonical_order = AllChem.CanonicalRankAtoms(mol)
+            for atom, can_idx in zip(mol.GetAtoms(), canonical_order):
+                atom_name = atom.GetSymbol().upper() + str(can_idx + 1)
+                if len(atom_name) > 4:
+                    msg = (
+                        f"{mol} has an atom with a name longer than "
+                        f"4 characters: {atom_name}."
+                    )
+                    raise ValueError(msg)
+                atom.SetProp("name", atom_name)
+
+            success = compute_3d_conformer(mol)
+            if not success:
+                msg = f"Failed to compute 3D conformer for {mol}"
+                raise ValueError(msg)
+
+            mol_no_h = AllChem.RemoveHs(mol, sanitize=False)
+            affinity_mw = AllChem.Descriptors.MolWt(mol_no_h) if affinity else None
+            extra_mols[f"LIG{ligand_id}"] = mol_no_h
+            residue = parse_ccd_residue(
+                name=f"LIG{ligand_id}",
+                ref_mol=mol,
+                res_idx=0,
+            )
+
+            ligand_id += 1
+            parsed_chain = ParsedChain(
+                entity=entity_id,
+                residues=[residue],
+                type=const.chain_type_ids["NONPOLYMER"],
+                cyclic_period=0,
+                sequence=None,
+                affinity=affinity,
+                affinity_mw=affinity_mw,
+            )
+
+            assert not items[0][entity_type].get(
+                "cyclic", False
+            ), "Cyclic flag is not supported for ligands"
+
         else:
             msg = f"Invalid entity type: {entity_type}"
             raise ValueError(msg)
@@ -1393,15 +1447,6 @@ def parse_boltz_schema(  # noqa: C901, PLR0915, PLR0912
                 chain_id=asym_id,
                 mw=chain.affinity_mw,
             )
-            # Save affinity info if output directory is specified
-            if output_dir is not None:
-                affinity_path = subfolder / "affinity_info.json"
-                with open(affinity_path, "w") as f:
-                    json.dump({
-                        "chain_id": asym_id,
-                        "mw": chain.affinity_mw,
-                        "chain_name": chain_name
-                    }, f)
 
         # Find all copies of this chain in the assembly
         entity_id = int(chain.entity)

boltz/main.py

@@ -498,13 +498,25 @@ def process_input(  # noqa: C901, PLR0912, PLR0915, D103
 ) -> None:
     try:
         # Parse data
-        if path.suffix in (".fa", ".fas", ".fasta"):
+        if path.is_dir():
+            # Process all YAML and FASTA files in the directory
+            targets = []
+            for file_path in path.glob("*"):
+                if file_path.suffix in (".fa", ".fas", ".fasta"):
+                    target = parse_fasta(file_path, ccd, mol_dir, boltz2)
+                    targets.append(target)
+                elif file_path.suffix in (".yml", ".yaml"):
+                    target = parse_yaml(file_path, ccd, mol_dir, boltz2)
+                    if not isinstance(target, list):
+                        target = [target]
+                    targets.extend(target)
+        elif path.suffix in (".fa", ".fas", ".fasta"):
             target = parse_fasta(path, ccd, mol_dir, boltz2)
+            targets = [target]
         elif path.suffix in (".yml", ".yaml"):
-            target = parse_yaml(path, ccd, mol_dir, boltz2)
-        elif path.is_dir():
-            msg = f"Found directory {path} instead of .fasta or .yaml, skipping."
-            raise RuntimeError(msg)  # noqa: TRY301
+            targets = parse_yaml(path, ccd, mol_dir, boltz2)
+            if not isinstance(targets, list):
+                targets = [targets]
         else:
             msg = (
                 f"Unable to parse filetype {path.suffix}, "
@@ -512,96 +524,98 @@ def process_input(  # noqa: C901, PLR0912, PLR0915, D103
             )
             raise RuntimeError(msg)  # noqa: TRY301
 
-        # Get target id
-        target_id = target.record.id
-
-        # Get all MSA ids and decide whether to generate MSA
-        to_generate = {}
-        prot_id = const.chain_type_ids["PROTEIN"]
-        for chain in target.record.chains:
-            # Add to generate list, assigning entity id
-            if (chain.mol_type == prot_id) and (chain.msa_id == 0):
-                entity_id = chain.entity_id
-                msa_id = f"{target_id}_{entity_id}"
-                to_generate[msa_id] = target.sequences[entity_id]
-                chain.msa_id = msa_dir / f"{msa_id}.csv"
-
-            # We do not support msa generation for non-protein chains
-            elif chain.msa_id == 0:
-                chain.msa_id = -1
-
-        # Generate MSA
-        if to_generate and not use_msa_server:
-            msg = "Missing MSA's in input and --use_msa_server flag not set."
-            raise RuntimeError(msg)  # noqa: TRY301
-
-        if to_generate:
-            msg = f"Generating MSA for {path} with {len(to_generate)} protein entities."
-            click.echo(msg)
-            compute_msa(
-                data=to_generate,
-                target_id=target_id,
-                msa_dir=msa_dir,
-                msa_server_url=msa_server_url,
-                msa_pairing_strategy=msa_pairing_strategy,
-            )
+        # Process each target
+        for target in targets:
+            # Get target id
+            target_id = target.record.id
+
+            # Get all MSA ids and decide whether to generate MSA
+            to_generate = {}
+            prot_id = const.chain_type_ids["PROTEIN"]
+            for chain in target.record.chains:
+                # Add to generate list, assigning entity id
+                if (chain.mol_type == prot_id) and (chain.msa_id == 0):
+                    entity_id = chain.entity_id
+                    msa_id = f"{target_id}_{entity_id}"
+                    to_generate[msa_id] = target.sequences[entity_id]
+                    chain.msa_id = msa_dir / f"{msa_id}.csv"
+
+                # We do not support msa generation for non-protein chains
+                elif chain.msa_id == 0:
+                    chain.msa_id = -1
+
+            # Generate MSA
+            if to_generate and not use_msa_server:
+                msg = "Missing MSA's in input and --use_msa_server flag not set."
+                raise RuntimeError(msg)  # noqa: TRY301
+
+            if to_generate:
+                msg = f"Generating MSA for {path} with {len(to_generate)} protein entities."
+                click.echo(msg)
+                compute_msa(
+                    data=to_generate,
+                    target_id=target_id,
+                    msa_dir=msa_dir,
+                    msa_server_url=msa_server_url,
+                    msa_pairing_strategy=msa_pairing_strategy,
+                )
 
-        # Parse MSA data
-        msas = sorted({c.msa_id for c in target.record.chains if c.msa_id != -1})
-        msa_id_map = {}
-        for msa_idx, msa_id in enumerate(msas):
-            # Check that raw MSA exists
-            msa_path = Path(msa_id)
-            if not msa_path.exists():
-                msg = f"MSA file {msa_path} not found."
-                raise FileNotFoundError(msg)  # noqa: TRY301
-
-            # Dump processed MSA
-            processed = processed_msa_dir / f"{target_id}_{msa_idx}.npz"
-            msa_id_map[msa_id] = f"{target_id}_{msa_idx}"
-            if not processed.exists():
-                # Parse A3M
-                if msa_path.suffix == ".a3m":
-                    msa: MSA = parse_a3m(
-                        msa_path,
-                        taxonomy=None,
-                        max_seqs=max_msa_seqs,
-                    )
-                elif msa_path.suffix == ".csv":
-                    msa: MSA = parse_csv(msa_path, max_seqs=max_msa_seqs)
-                else:
-                    msg = f"MSA file {msa_path} not supported, only a3m or csv."
-                    raise RuntimeError(msg)  # noqa: TRY301
-
-                msa.dump(processed)
-
-        # Modify records to point to processed MSA
-        for c in target.record.chains:
-            if (c.msa_id != -1) and (c.msa_id in msa_id_map):
-                c.msa_id = msa_id_map[c.msa_id]
-
-        # Dump templates
-        for template_id, template in target.templates.items():
-            name = f"{target.record.id}_{template_id}.npz"
-            template_path = processed_templates_dir / name
-            template.dump(template_path)
-
-        # Dump constraints
-        constraints_path = processed_constraints_dir / f"{target.record.id}.npz"
-        target.residue_constraints.dump(constraints_path)
-
-        # Dump extra molecules
-        Chem.SetDefaultPickleProperties(Chem.PropertyPickleOptions.AllProps)
-        with (processed_mols_dir / f"{target.record.id}.pkl").open("wb") as f:
-            pickle.dump(target.extra_mols, f)
-
-        # Dump structure
-        struct_path = structure_dir / f"{target.record.id}.npz"
-        target.structure.dump(struct_path)
-
-        # Dump record
-        record_path = records_dir / f"{target.record.id}.json"
-        target.record.dump(record_path)
+            # Parse MSA data
+            msas = sorted({c.msa_id for c in target.record.chains if c.msa_id != -1})
+            msa_id_map = {}
+            for msa_idx, msa_id in enumerate(msas):
+                # Check that raw MSA exists
+                msa_path = Path(msa_id)
+                if not msa_path.exists():
+                    msg = f"MSA file {msa_path} not found."
+                    raise FileNotFoundError(msg)  # noqa: TRY301
+
+                # Dump processed MSA
+                processed = processed_msa_dir / f"{target_id}_{msa_idx}.npz"
+                msa_id_map[msa_id] = f"{target_id}_{msa_idx}"
+                if not processed.exists():
+                    # Parse A3M
+                    if msa_path.suffix == ".a3m":
+                        msa: MSA = parse_a3m(
+                            msa_path,
+                            taxonomy=None,
+                            max_seqs=max_msa_seqs,
+                        )
+                    elif msa_path.suffix == ".csv":
+                        msa: MSA = parse_csv(msa_path, max_seqs=max_msa_seqs)
+                    else:
+                        msg = f"MSA file {msa_path} not supported, only a3m or csv."
+                        raise RuntimeError(msg)  # noqa: TRY301
+
+                    msa.dump(processed)
+
+            # Modify records to point to processed MSA
+            for c in target.record.chains:
+                if (c.msa_id != -1) and (c.msa_id in msa_id_map):
+                    c.msa_id = msa_id_map[c.msa_id]
+
+            # Dump templates
+            for template_id, template in target.templates.items():
+                name = f"{target.record.id}_{template_id}.npz"
+                template_path = processed_templates_dir / name
+                template.dump(template_path)
+
+            # Dump constraints
+            constraints_path = processed_constraints_dir / f"{target.record.id}.npz"
+            target.residue_constraints.dump(constraints_path)
+
+            # Dump extra molecules
+            Chem.SetDefaultPickleProperties(Chem.PropertyPickleOptions.AllProps)
+            with (processed_mols_dir / f"{target.record.id}.pkl").open("wb") as f:
+                pickle.dump(target.extra_mols, f)
+
+            # Dump structure
+            struct_path = structure_dir / f"{target.record.id}.npz"
+            target.structure.dump(struct_path)
+
+            # Dump record
+            record_path = records_dir / f"{target.record.id}.json"
+            target.record.dump(record_path)
 
     except Exception as e:  # noqa: BLE001
         import traceback

{boltz_vsynthes-1.0.38.dist-info → boltz_vsynthes-1.0.40.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: boltz-vsynthes
-Version: 1.0.38
+Version: 1.0.40
 Summary: Boltz for VSYNTHES
 Requires-Python: <3.13,>=3.10
 Description-Content-Type: text/markdown

{boltz_vsynthes-1.0.38.dist-info → boltz_vsynthes-1.0.40.dist-info}/RECORD

@@ -1,5 +1,5 @@
 boltz/__init__.py,sha256=F_-so3S40iZrSZ89Ge4TS6aZqwWyZXq_H4AXGDlbA_g,187
-boltz/main.py,sha256=i5_15JZ9vjZ9RSLZb2F0a7scuQ0QfFkgUQVftTiD3h0,39945
+boltz/main.py,sha256=SHM-t-9wjwjTJmWR4N5SrAHxk2vgz7fTruz5shiixVc,40882
 boltz/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/data/const.py,sha256=1M-88Z6HkfKY6MkNtqcj3b9P-oX9xEXluh3qM_u8dNU,26779
 boltz/data/mol.py,sha256=maOpPHEGX1VVXCIFY6pQNGF7gUBZPAfgSvuPf2QO1yc,34268
@@ -40,7 +40,7 @@ boltz/data/parse/mmcif.py,sha256=25kEXCkx-OuaawAs7cdz0fxdRu5_CCO0AV00u84PrjQ,368
 boltz/data/parse/mmcif_with_constraints.py,sha256=WHYZckSqUwu-Nb9vmVmxHmC7uxwVrF7AVUeVKsc5wGQ,51473
 boltz/data/parse/pdb.py,sha256=iybk4p2UgUy_ABGprDq_xxyPSdm1HAZsGTM0lhxVEwM,1654
 boltz/data/parse/pdb_download.py,sha256=wge-scX-lOatX0q83W1wOsaql99rYp-6uGWSHEc995M,2718
-boltz/data/parse/schema.py,sha256=b0Mh1eCg6gTyOQt7GkEFAQdYCZJ1jqAJbUy9Tv53K4E,64781
+boltz/data/parse/schema.py,sha256=p4KIAVzQAuApcxRLHc6-KKG7ICgLmEWVzE8Qqm6v04w,66402
 boltz/data/parse/sdf.py,sha256=fs3MQVClDcCzxJaeVYiDuoh-fUrYc8Tcd5Bz8ws3FKI,2052
 boltz/data/parse/yaml.py,sha256=M3dRQK2mMDue3bPSO_T2ThaVojSMrOV7rMY-KXQvaGQ,2047
 boltz/data/sample/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -107,9 +107,9 @@ boltz/model/optim/scheduler.py,sha256=nB4jz0CZ4pR4n08LQngExL_pNycIdYI8AXVoHPnZWQ
 boltz/model/potentials/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/model/potentials/potentials.py,sha256=vev8Vjfs-ML1hyrdv_R8DynG4wSFahJ6nzPWp7CYQqw,17507
 boltz/model/potentials/schedules.py,sha256=m7XJjfuF9uTX3bR9VisXv1rvzJjxiD8PobXRpcBBu1c,968
-boltz_vsynthes-1.0.38.dist-info/licenses/LICENSE,sha256=8GZ_1eZsUeG6jdqgJJxtciWzADfgLEV4LY8sKUOsJhc,1102
-boltz_vsynthes-1.0.38.dist-info/METADATA,sha256=HtZ8GekM9xBSPv24CUhVnarUX_GnVP-_tvIM8HfLCZc,7171
-boltz_vsynthes-1.0.38.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-boltz_vsynthes-1.0.38.dist-info/entry_points.txt,sha256=n5a5I35ntu9lmyr16oZgHPFY0b0YxjiixY7m7nbMTLc,41
-boltz_vsynthes-1.0.38.dist-info/top_level.txt,sha256=MgU3Jfb-ctWm07YGMts68PMjSh9v26D0gfG3dFRmVFA,6
-boltz_vsynthes-1.0.38.dist-info/RECORD,,
+boltz_vsynthes-1.0.40.dist-info/licenses/LICENSE,sha256=8GZ_1eZsUeG6jdqgJJxtciWzADfgLEV4LY8sKUOsJhc,1102
+boltz_vsynthes-1.0.40.dist-info/METADATA,sha256=z2kizv_5w3PrpKHsDV_GXjhzQDRxRCWWT2pOESvbcFU,7171
+boltz_vsynthes-1.0.40.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+boltz_vsynthes-1.0.40.dist-info/entry_points.txt,sha256=n5a5I35ntu9lmyr16oZgHPFY0b0YxjiixY7m7nbMTLc,41
+boltz_vsynthes-1.0.40.dist-info/top_level.txt,sha256=MgU3Jfb-ctWm07YGMts68PMjSh9v26D0gfG3dFRmVFA,6
+boltz_vsynthes-1.0.40.dist-info/RECORD,,