2

I'm working with RDKit and have the following issue. I'm trying to create a function that encode a molecule from a SMILES string into a fingerprint. But an ArgumentError occurs and I can't figure it out.

This is my code:

 def get_fp(dfx, method="maccs", n_bits=2048):

    ligands = [Chem.MolFromSmiles(mol) for mol in dfx.canonical_smiles]
    
    if method == "morgan2":
        data1 = []
        for mol in ligands:
            ecfp4 = [int(x) for x in AllChem.GetMorganFingerprintAsBitVect(mol, 2, n_Bits=n_bits)]
            data1.append(ecfp4)
        
        ecfp4_df = pd.DataFrame(data1, index=dfx.index)    
        return ecfp4_df
    
    if method == "morgan3":
        data2 = []
        for mol in ligands:
            ecfp6 = [int(x) for x in AllChem.GetMorganFingerprintAsBitVect(mol, 3, n_Bits=n_bits)]
            data2.append(ecfp6)
        ecfp6_df = pd.DataFrame(data2, index=dfx.index)
        return ecfp6_df
    
    if method == "maccs":
        data3 = []
        for mol in ligands:
            maccs = [int(x) for x in MACCSkeys.GenMACCSKeys(mol)]
            data3.append(maccs)
        maccs_df = pd.DataFrame(data3, index=dfx.index)
        return maccs_df

For "maccs" method it worked but for morgan didn't. I think it is a function problem but I cannot debug it... This is the error:

ArgumentError: Python argument types in rdkit.Chem.rdMolDescriptors.GetMorganFingerprintAsBitVect(Mol, int) did not match C++ signature: GetMorganFingerprintAsBitVect(class RDKit::ROMol mol, unsigned int radius, unsigned int nBits=2048, class boost::python::api::object invariants=[], class boost::python::api::object fromAtoms=[], bool useChirality=False, bool useBondTypes=True, bool useFeatures=False, class boost::python::api::object bitInfo=None, bool includeRedundantEnvironments=False)

Thanks

SEBASTIAN
  • 75
  • 1
  • 8

1 Answers1

1

It is just a typo.

Change

AllChem.GetMorganFingerprintAsBitVect(mol, 2, n_Bits = n_bits)

to

AllChem.GetMorganFingerprintAsBitVect(mol, 2, nBits = n_bits)

rapelpy
  • 1,684
  • 1
  • 11
  • 14