Re: [Rdkit-discuss] MolFromMol2Block changes carboxylic group representation

Hi Maria I would say that the behaviour of RDKit with your MOL2 file is right. SMILES notation doesn't have a way to represent the delocalised form of a carboxylate anion, so the O=C[O-] form is the correct SMILES for this structure. RDKit does a good job in recognising that it's the anion based

Re: [Rdkit-discuss] MolFromMol2Block changes carboxylic group representation

Hi Maria, This is one of many routines molecule undergoes when reading from a "Corina Mol2", which is implemented in RDKit. Unfortunately there is no way to turn it off, due to the (gu)estimation of formal charges. Pozdrawiam, | Best regards, Maciek Wójcikowski mac...@wojcikowski.pl

Re: [Rdkit-discuss] MolFromMol2Block changes carboxylic group representation

Sorry, i forgot to ask a question: how can i retain the initial form? Maria 2018-03-28 11:46 GMT+02:00 Maria Matveyeva : > Hello all, > When i read mol2 format from either mol2 file or block, it changes > represenation of carboxylic group from tripos aromatic

[Rdkit-discuss] MolFromMol2Block changes carboxylic group representation

Hello all, When i read mol2 format from either mol2 file or block, it changes represenation of carboxylic group from tripos aromatic representaiton with -0.5 charges on oxygens to representaion with one single and one double bond (when same representation read from sdf/mol it retains "aromatic"