Author: jp

Cheminformatics is hard. That is a great quote from Prof. Paul Finn. I think part of it is due to the nature of chemistry (e.g. which is the correct tautomer for this molecule?), and part of it is because of the lack of “standard” process definitions.

So I am revisiting the standardization (of the molecule)/normalization(of functional groups) pipeline for ML, and I had to post to the extremely helpful RDKit mailing list for help (here). Using the excellent sources they pointed to me, I ended up with the following (which will surely come in handy in a few months time when I go through the whole process again):

def standardize(smiles):
    # follows the steps in
    # https://github.com/greglandrum/RSC_OpenScience_Standardization_202104/blob/main/MolStandardize%20pieces.ipynb
    # as described **excellently** (by Greg) in
    # https://www.youtube.com/watch?v=eWTApNX8dJQ
    mol = Chem.MolFromSmiles(smiles)
    
    # removeHs, disconnect metal atoms, normalize the molecule, reionize the molecule
    clean_mol = rdMolStandardize.Cleanup(mol) 
    
    # if many fragments, get the "parent" (the actual mol we are interested in) 
    parent_clean_mol = rdMolStandardize.FragmentParent(clean_mol)
        
    # try to neutralize molecule
    uncharger = rdMolStandardize.Uncharger() # annoying, but necessary as no convenience method exists
    uncharged_parent_clean_mol = uncharger.uncharge(parent_clean_mol)
    
    # note that no attempt is made at reionization at this step
    # nor at ionization at some pH (rdkit has no pKa caculator)
    # the main aim to to represent all molecules from different sources
    # in a (single) standard way, for use in ML, catalogue, etc.
    
    te = rdMolStandardize.TautomerEnumerator() # idem
    taut_uncharged_parent_clean_mol = te.Canonicalize(uncharged_parent_clean_mol)
    
    return taut_uncharged_parent_clean_mol

I am always sending the same canned response to students who would like to do an FYP or a dissertation with me on the subjects I dabble in, Computer-Aided Drug Design (Discovery), Virtual Screening (VS), Ligand-based and Structure-Based methods, Cheminformatics, Bioinformatics and Computational Chemistry. Perhaps the first step for any student is to realize the hierarchy of these fields (and the differences between them). I am including a reading list – which helps you  bootstrap the subject, and hopefully helps you determine if this is really something for you. The jargon will be daunting at first (especially if you are a computer scientist), but that is only an initial hurdle and hopefully you get familiar with the big words quickly. You do not need to understand everything, you just need to understand enough. Remember brick walls are there to show us how badly we want things! (Watch this: long and touching).