feat: added versioning, HOSE codes and license details to the swagger docs

app/main.py

@@ -1,13 +1,25 @@
 from fastapi import FastAPI
 from fastapi.responses import RedirectResponse
+from fastapi_versioning import VersionedFastAPI
 
 # from .config import settings
-from .routers import chem, converters, compose, decimer
+from .routers import chem, converters, decimer
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.openapi.utils import get_openapi
-import os
 
-app = FastAPI()
+app = FastAPI(
+    title="Cheminf Micro Services",
+    description="This set of essential and valuable microservices is designed to be accessed via API calls to support cheminformatics. Generally, it is designed to work with SMILES-based inputs and could be used to translate between different machine-readable representations, get Natural Product (NP) likeliness scores, visualize chemical structures, and generate descriptors. In addition, the microservices also host an instance of STOUT and another instance of DECIMER (two deep learning models for IUPAC name generation and optical chemical structure recognition, respectively).",
+    terms_of_service="https://github.com/Steinbeck-Lab",
+    contact={
+        "name": "Steinbeck Lab",
+        "url": "https://cheminf.uni-jena.de/",
+        "email": "caffeine@listserv.uni-jena.de",
+    },
+    license_info={
+        "name": "CC BY 4.0",
+        "url": "https://creativecommons.org/licenses/by/4.0/",
+    },
+)
 
 origins = ["*"]
 
@@ -21,7 +33,6 @@
 
 app.include_router(chem.router)
 app.include_router(converters.router)
-app.include_router(compose.router)
 app.include_router(decimer.router)
 
 
@@ -30,20 +41,18 @@ async def docs_redirect():
     return RedirectResponse(url="/docs")
 
 
-def custom_openapi():
-    if app.openapi_schema:
-        return app.openapi_schema
-    openapi_schema = get_openapi(
-        title="Cheminf Micro Services",
-        version=os.getenv("RELEASE_VERSION", "latest"),
-        description="This set of essential and valuable microservices is designed to be accessed via API calls to support cheminformatics. Generally, it is designed to work with SMILES-based inputs and could be used to translate between different machine-readable representations, get Natural Product (NP) likeliness scores, visualize chemical structures, and generate descriptors. In addition, the microservices also host an instance of STOUT and another instance of DECIMER (two deep learning models for IUPAC name generation and optical chemical structure recognition, respectively).",
-        routes=app.routes,
-    )
-    openapi_schema["info"]["x-logo"] = {
-        "url": "https://github.com/Steinbeck-Lab/cheminformatics-python-microservice/raw/main/public/img/logo.png"
-    }
-    app.openapi_schema = openapi_schema
-    return app.openapi_schema
-
-
-app.openapi = custom_openapi
+app = VersionedFastAPI(
+    app,
+    version_format="{major}",
+    prefix_format="/v{major}",
+    terms_of_service="https://github.com/Steinbeck-Lab",
+    contact={
+        "name": "Steinbeck Lab",
+        "url": "https://cheminf.uni-jena.de/",
+        "email": "caffeine@listserv.uni-jena.de",
+    },
+    license_info={
+        "name": "CC BY 4.0",
+        "url": "https://creativecommons.org/licenses/by/4.0/",
+    },
+)

app/modules/cdkmodules.py

@@ -394,3 +394,27 @@ def getCXSMILES(smiles: str):
     )
     CXSMILES = SmilesGenerator.create(moleculeSDG)
     return str(CXSMILES)
+
+
+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.
+    Args:
+            smiles (string): SMILES string given by the user.
+    Returns:
+            mol object (string): CDK Structure Diagram Layout mol block.
+    """
+    if any(char.isspace() for char in smiles):
+        smiles = smiles.replace(" ", "+")
+    SCOB = JClass(cdk_base + ".silent.SilentChemObjectBuilder")
+    SmilesParser = JClass(cdk_base + ".smiles.SmilesParser")(SCOB.getInstance())
+    molecule = SmilesParser.parseSmiles(smiles)
+    HOSECodeGenerator = JClass(cdk_base + ".tools.HOSECodeGenerator")()
+    HOSECodes = []
+    atoms = molecule.atoms()
+    for atom in atoms:
+        moleculeHOSECode = HOSECodeGenerator.getHOSECode(
+            molecule, atom, noOfSpheres, ringsize
+        )
+        HOSECodes.append(str(moleculeHOSECode))
+    return HOSECodes

app/modules/depict.py

@@ -51,7 +51,6 @@ def getCDKDepiction(
         moleculeSDG = getCDKSDG(smiles)
 
     if moleculeSDG:
-
         # Rotate molecule
         point = JClass(cdk_base + ".geometry.GeometryTools").get2DCenter(moleculeSDG)
         JClass(cdk_base + ".geometry.GeometryTools").rotate(

app/modules/rdkitmodules.py

@@ -2,6 +2,7 @@
 from rdkit import Chem, DataStructs
 from rdkit.Chem import AllChem, Descriptors, QED, Lipinski, rdMolDescriptors, rdmolops
 from app.modules.cdkmodules import getCDKSDGMol
+from hosegen import HoseGenerator
 
 
 def checkSMILES(smiles: str):
@@ -134,3 +135,15 @@ def getTanimotoSimilarityRDKit(smiles1, smiles2):
     similarity = DataStructs.TanimotoSimilarity(fp1, fp2)
 
     return similarity
+
+
+async def getRDKitHOSECodes(smiles: str, noOfSpheres: int):
+    if any(char.isspace() for char in smiles):
+        smiles = smiles.replace(" ", "+")
+    mol = Chem.MolFromSmiles(smiles)
+    gen = HoseGenerator()
+    hosecodes = []
+    for i in range(0, len(mol.GetAtoms()) - 1):
+        hosecode = gen.get_Hose_codes(mol, i, noOfSpheres)
+        hosecodes.append(hosecode)
+    return hosecodes

app/routers/chem.py

@@ -9,9 +9,17 @@
 from fastapi.responses import Response, HTMLResponse
 from app.modules.npscorer import getNPScore
 from app.modules.classyfire import classify, result
-from app.modules.cdkmodules import getCDKSDGMol, getTanimotoSimilarityCDK
+from app.modules.cdkmodules import (
+    getCDKSDGMol,
+    getTanimotoSimilarityCDK,
+    getCDKHOSECodes,
+)
 from app.modules.depict import getRDKitDepiction, getCDKDepiction
-from app.modules.rdkitmodules import get3Dconformers, getTanimotoSimilarityRDKit
+from app.modules.rdkitmodules import (
+    get3Dconformers,
+    getTanimotoSimilarityRDKit,
+    getRDKitHOSECodes,
+)
 from app.modules.coconutdescriptors import getCOCONUTDescriptors
 from app.modules.alldescriptors import getTanimotoSimilarity
 import pandas as pd
@@ -297,6 +305,17 @@ async def Depict3D_Molecule(
         return templates.TemplateResponse("mol.html", content)
 
 
+@router.get("/hosecode")
+async def HOSE_Codes(framework: str, smiles: str, spheres: int, ringsize: bool = False):
+    if smiles:
+        if framework == "cdk":
+            return await getCDKHOSECodes(smiles, spheres, ringsize)
+        elif framework == "rdkit":
+            return await getRDKitHOSECodes(smiles, spheres)
+    else:
+        return "Error reading SMILES string, check again."
+
+
 # @app.get("/molecules/", response_model=List[schemas.Molecule])
 # def read_molecules(skip: int = 0, limit: int = 100, db: Session = Depends(get_db)):
 #     molecules = crud.get_molecules(db, skip=skip, limit=limit)

app/routers/decimer.py

@@ -24,7 +24,7 @@ async def DECIMER_Index():
 @router.post("/process")
 async def Extract_ChemicalInfo(
     path: Annotated[str, Body(embed=True)] = None,
-    reference: Annotated[str, Body(embed=True)]  = None,
+    reference: Annotated[str, Body(embed=True)] = None,
     img: Annotated[str, Body(embed=True)] = None,
 ):
     """

requirements.txt

@@ -1,6 +1,7 @@
 fastapi>=0.80.0
 uvicorn>=0.15.0,<0.16.0
 fastapi-pagination==0.10.0
+fastapi-versioning>=0.10.0
 rdkit-pypi>=2022.09.4
 websockets==10.4
 pillow==9.4.0
@@ -18,3 +19,4 @@ pillow-heif==0.10.0
 selfies>=2.1.1
 jinja2
 pandas
+HOSE_code_generator @ git+https://github.com/Ratsemaat/HOSE_code_generator