boltz-vsynthes 1.0.9__py3-none-any.whl → 1.0.11__py3-none-any.whl

boltz/data/mol.py

@@ -30,10 +30,6 @@ def load_molecules(moldir: str, molecules: list[str]) -> dict[str, Mol]:
     """
     loaded_mols = {}
     for molecule in molecules:
-        # Skip if it's a SMILES string (starts with LIG)
-        if molecule.startswith("LIG"):
-            continue
-            
         path = Path(moldir) / f"{molecule}.pkl"
         if not path.exists():
             msg = f"CCD component {molecule} not found!"

boltz/data/parse/__init__.py

@@ -0,0 +1,21 @@
+from boltz.data.parse.pdb import parse_pdb
+from boltz.data.parse.sdf import parse_sdf
+from boltz.data.parse.pdb_download import parse_pdb_id
+from boltz.data.parse.yaml import parse_yaml
+from boltz.data.parse.fasta import parse_fasta
+from boltz.data.parse.a3m import parse_a3m
+from boltz.data.parse.csv import parse_csv
+from boltz.data.parse.mmcif import parse_mmcif
+from boltz.data.parse.mmcif_with_constraints import parse_mmcif_with_constraints
+
+__all__ = [
+    "parse_pdb",
+    "parse_sdf",
+    "parse_pdb_id",
+    "parse_yaml",
+    "parse_fasta",
+    "parse_a3m",
+    "parse_csv",
+    "parse_mmcif",
+    "parse_mmcif_with_constraints",
+]

boltz/data/parse/pdb.py

@@ -0,0 +1,72 @@
+from pathlib import Path
+from typing import Optional
+
+from Bio import PDB
+from Bio.PDB.Polypeptide import three_to_one
+from Bio.Data.IUPACData import protein_letters_3to1
+from rdkit import Chem
+from rdkit.Chem.rdchem import Mol
+
+from boltz.data.types import Target
+from boltz.data.parse.yaml import parse_boltz_schema
+
+
+def parse_pdb(
+    path: Path,
+    ccd: dict[str, Mol],
+    mol_dir: Path,
+    boltz2: bool = False,
+) -> Target:
+    """Parse a PDB file.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the PDB file.
+    ccd : Dict
+        Dictionary of CCD components.
+    mol_dir : Path
+        Path to the directory containing the molecules.
+    boltz2 : bool
+        Whether to parse the input for Boltz2.
+
+    Returns
+    -------
+    Target
+        The parsed target.
+    """
+    # Read PDB file
+    parser = PDB.PDBParser(QUIET=True)
+    structure = parser.get_structure("protein", str(path))
+
+    # Convert to yaml format
+    sequences = []
+    for model in structure:
+        for chain in model:
+            # Get chain sequence
+            seq = ""
+            for residue in chain:
+                if residue.id[0] == " ":  # Only standard residues
+                    try:
+                        seq += protein_letters_3to1[residue.resname]
+                    except KeyError:
+                        continue
+
+            if seq:  # Only add if sequence is not empty
+                molecule = {
+                    "protein": {
+                        "id": chain.id,
+                        "sequence": seq,
+                        "modifications": [],
+                    },
+                }
+                sequences.append(molecule)
+
+    data = {
+        "sequences": sequences,
+        "bonds": [],
+        "version": 1,
+    }
+
+    name = path.stem
+    return parse_boltz_schema(name, data, ccd, mol_dir, boltz2) 

boltz/data/parse/pdb_download.py

@@ -0,0 +1,114 @@
+import os
+from pathlib import Path
+from typing import Optional
+
+import requests
+from Bio import PDB
+from Bio.Data.IUPACData import protein_letters_3to1
+from rdkit import Chem
+from rdkit.Chem.rdchem import Mol
+
+from boltz.data.types import Target
+from boltz.data.parse.yaml import parse_boltz_schema
+
+
+def download_pdb(pdb_id: str, cache_dir: Path) -> Path:
+    """Download a PDB file by ID.
+
+    Parameters
+    ----------
+    pdb_id : str
+        The PDB ID to download.
+    cache_dir : Path
+        The directory to cache the downloaded file.
+
+    Returns
+    -------
+    Path
+        The path to the downloaded PDB file.
+    """
+    # Create cache directory if it doesn't exist
+    cache_dir.mkdir(parents=True, exist_ok=True)
+    
+    # Check if file already exists in cache
+    pdb_path = cache_dir / f"{pdb_id.lower()}.pdb"
+    if pdb_path.exists():
+        return pdb_path
+    
+    # Download from RCSB PDB
+    url = f"https://files.rcsb.org/download/{pdb_id.lower()}.pdb"
+    response = requests.get(url, stream=True)
+    response.raise_for_status()
+    
+    # Save to cache
+    with pdb_path.open("wb") as f:
+        for chunk in response.iter_content(chunk_size=8192):
+            f.write(chunk)
+    
+    return pdb_path
+
+
+def parse_pdb_id(
+    pdb_id: str,
+    ccd: dict[str, Mol],
+    mol_dir: Path,
+    cache_dir: Path,
+    boltz2: bool = False,
+) -> Target:
+    """Parse a PDB file by ID.
+
+    Parameters
+    ----------
+    pdb_id : str
+        The PDB ID to parse.
+    ccd : Dict
+        Dictionary of CCD components.
+    mol_dir : Path
+        Path to the directory containing the molecules.
+    cache_dir : Path
+        The directory to cache downloaded PDB files.
+    boltz2 : bool
+        Whether to parse the input for Boltz2.
+
+    Returns
+    -------
+    Target
+        The parsed target.
+    """
+    # Download PDB file
+    pdb_path = download_pdb(pdb_id, cache_dir)
+    
+    # Read PDB file
+    parser = PDB.PDBParser(QUIET=True)
+    structure = parser.get_structure("protein", str(pdb_path))
+
+    # Convert to yaml format
+    sequences = []
+    for model in structure:
+        for chain in model:
+            # Get chain sequence
+            seq = ""
+            for residue in chain:
+                if residue.id[0] == " ":  # Only standard residues
+                    try:
+                        seq += protein_letters_3to1[residue.resname]
+                    except KeyError:
+                        continue
+
+            if seq:  # Only add if sequence is not empty
+                molecule = {
+                    "protein": {
+                        "id": chain.id,
+                        "sequence": seq,
+                        "modifications": [],
+                    },
+                }
+                sequences.append(molecule)
+
+    data = {
+        "sequences": sequences,
+        "bonds": [],
+        "version": 1,
+    }
+
+    return parse_boltz_schema(pdb_id, data, ccd, mol_dir, boltz2)