Merge pull request #197 from Steinbeck-Lab/dev-kohulan

feat: Add openbabel modules #185,#186 and #195

Dockerfile

@@ -2,6 +2,7 @@ FROM continuumio/miniconda3 AS cheminf-python-ms
 
 ENV PYTHON_VERSION=3.10
 ENV RDKIT_VERSION=2023.03.1
+ENV OPENBABEL_VERSION=v3.1
 
 ARG RELEASE_VERSION
 ENV RELEASE_VERSION=${RELEASE_VERSION}
@@ -12,8 +13,9 @@ RUN apt-get update && \
     apt-get update -y && \
     apt-get install -y openjdk-11-jre
 
-RUN conda install -c conda-forge python>=$PYTHON_VERSION
+RUN conda install -c conda-forge python>=PYTHON_VERSION
 RUN conda install -c conda-forge rdkit>=RDKIT_VERSION
+RUN conda install -c conda-forge openbabel>=OPENBABEL_VERSION
 
 RUN python3 -m pip install -U pip
 

app/modules/toolkits/cdkmodules.py

@@ -401,6 +401,46 @@ def getCXSMILES(smiles: str):
     return str(CXSMILES)
 
 
+def getCanonSMILES(smiles: str):
+    """This function takes the user input SMILES and creates a
+    Canonical SMILES string with 2D atom coordinates
+    Args:
+            smiles (string): SMILES string given by the user.
+    Returns:
+            smiles (string): Canonical SMILES string.
+
+    """
+    moleculeSDG = getCDKSDG(smiles)
+    SmiFlavor = JClass(cdk_base + ".smiles.SmiFlavor")
+    SmilesGenerator = JClass(cdk_base + ".smiles.SmilesGenerator")(SmiFlavor.Absolute)
+    CanonicalSMILES = SmilesGenerator.create(moleculeSDG)
+    return str(CanonicalSMILES)
+
+
+def getInChI(smiles: str, InChIKey=False):
+    """This function takes the user input SMILES and creates a
+    InChI string
+    Args:
+            smiles (string): SMILES string given by the user.
+    Returns:
+            smiles (string): InChI/InChIKey string.
+
+    """
+    moleculeSDG = getCDKSDG(smiles)
+    InChIGeneratorFactory = JClass(cdk_base + ".inchi.InChIGeneratorFactory")
+    InChI = (
+        InChIGeneratorFactory.getInstance().getInChIGenerator(moleculeSDG).getInchi()
+    )
+    if InChIKey:
+        InChIKey = (
+            InChIGeneratorFactory.getInstance()
+            .getInChIGenerator(moleculeSDG)
+            .getInchiKey()
+        )
+        return InChIKey
+    return InChI
+
+
 async def getCDKHOSECodes(smiles: str, noOfSpheres: int, ringsize: bool):
     """This function takes the user input SMILES and returns a mol
        block as a string with Structure Diagram Layout.

app/modules/toolkits/openbabelmodules.py

@@ -0,0 +1,82 @@
+from openbabel import openbabel as ob
+from openbabel import pybel
+
+def getOBCanonicalSMILES(smiles:str):
+    """This function takes an input as a SMILES string and
+    returns a Canonical SMILES.
+    Args (str): SMILES string.
+    Returns (str): Canonical SMILES string.
+    """
+    if any(char.isspace() for char in smiles):
+        smiles = smiles.replace(" ", "+")
+
+    # Create an Open Babel molecule object
+    mol = ob.OBMol()
+
+    conv = ob.OBConversion()
+    conv.SetInAndOutFormats("smi", "can")
+    conv.ReadString(mol, smiles)
+
+    canSMILES = conv.WriteString(mol)
+    canSMILES = canSMILES.strip()  # Remove leading/trailing whitespace
+    return canSMILES   
+
+
+def getOBInChI(smiles:str,InChIKey:bool=False):
+    """This function takes an input as a SMILES string and
+    returns a InChI
+    Args (str): SMILES string.
+    Returns (str): InChI string.
+    """
+    if any(char.isspace() for char in smiles):
+        smiles = smiles.replace(" ", "+")
+
+    # Create an Open Babel molecule object
+    mol = ob.OBMol()
+
+    # Create OBConversion
+    conv = ob.OBConversion()
+    conv.SetInAndOutFormats("smi", "inchi")
+    conv.ReadString(mol, smiles)
+
+    inchi = conv.WriteString(mol)
+    inchi = inchi.strip()  # Remove leading/trailing whitespace
+    if InChIKey:
+        conv.SetOptions("K", conv.OUTOPTIONS)
+        inchikey_ = conv.WriteString(mol).rstrip()
+        return inchikey_
+    return inchi
+
+def getOBMol(smiles:str, threeD:bool=False):
+    """This function takes an input as a SMILES string and
+    returns a 2D/3D mol block.
+    Args (str): SMILES string.
+    Returns (str): Mol block (2D/3D).
+    """
+    if any(char.isspace() for char in smiles):
+        smiles = smiles.replace(" ", "+")
+
+    if threeD:
+        mol = pybel.readstring("smi", smiles)
+        mol.addh()
+        mol.make3D()
+        mol.removeh()
+        return mol.write("mol")
+
+    # Create an Open Babel molecule object
+    mol = ob.OBMol()
+
+    conv = ob.OBConversion()
+    conv.SetInAndOutFormats("smi", "mol")
+    conv.ReadString(mol, smiles)
+
+    # Generate 2D coordinates
+    obBuilder = ob.OBBuilder()
+    obBuilder.Build(mol)    
+
+    mol_block = conv.WriteString(mol)
+    mol_block = mol_block.strip()  # Remove leading/trailing whitespace
+    return mol_block
+
+
+

app/modules/toolkits/rdkitmodules.py

@@ -194,3 +194,21 @@ def has_stereochemistry(smiles: str):
             return True
 
     return False
+
+
+def get2Dmol(smiles: str):
+    """This function takes an input as a SMILES string and
+    returns a 2D mol block.
+    Args (str): SMILES string.
+    Returns (str): 2D Mol block.
+    """
+    if any(char.isspace() for char in smiles):
+        smiles = smiles.replace(" ", "+")
+    mol = Chem.MolFromSmiles(smiles)
+
+    if mol:
+        AllChem.Compute2DCoords(mol)
+        molfile = Chem.MolToMolBlock(mol)
+        return molfile
+    else:
+        return "Error reading SMILES string, check again."

app/routers/chem.py

@@ -10,7 +10,6 @@
 from app.modules.npscorer import getNPScore
 from app.modules.classyfire import classify, result
 from app.modules.toolkits.cdkmodules import (
-    getCDKSDGMol,
     getTanimotoSimilarityCDK,
     getCDKHOSECodes,
 )
@@ -149,24 +148,6 @@ async def ClassyFire_result(id: str):
         return data
 
 
-@router.get("/cdk2d")
-async def CDK2D_Coordinates(smiles: str):
-    """
-    Generate 2D Coordinates using the CDK Structure diagram generator and return the mol block.
-
-    - **SMILES**: required (query)
-    """
-    if smiles:
-        mol = Chem.MolFromSmiles(smiles)
-        if mol:
-            return Response(
-                content=getCDKSDGMol(smiles).replace("$$$$\n", ""),
-                media_type="text/plain",
-            )
-        else:
-            return "Error reading SMILES string, check again."
-
-
 @router.get("/rdkit3d")
 async def RDKit3D_Mol(smiles: str):
     """