ChEMBL extension in credoscript

credoscript-chembl-activitycliffs.py

from credoscript.extensions import chembl

m = ma.fetch_by_molregno(410891)

# get all activity cliffs of this molecule
m.get_activity_cliffs(chembl.ActivityCliff.sali>10)

>>> [<ActivityCliff(455590, 2030543, 2030536)>,
     <ActivityCliff(455592, 2030585, 2030582)>,
     <ActivityCliff(455594, 2030630, 2030623)>,
     <ActivityCliff(455591, 2030571, 2030564)>,
     <ActivityCliff(455590, 2030543, 2030535)>,
     <ActivityCliff(455592, 2030585, 2030581)>,
     <ActivityCliff(455591, 2030571, 2030563)>]

# get the activity cliff with the highest SALI index
cliff = m.get_activity_cliffs(chembl.ActivityCliff.sali>10)[0]

cliff.delta_activity, cliff.delta_tanimoto, cliff.sali
>>> (1.44235914846045, 0.0517241379310345, 27.8856102035687)

cliff.ActivityBgn.standard_value, cliff.ActivityBgn.standard_units
>>> (13.0, 'nM')

cliff.ActivityEnd.standard_value, cliff.ActivityEnd.standard_units
>>> (360.0, 'nM')

cliff.MoleculeBgn.Structure.canonical_smiles, cliff.MoleculeEnd.Structure.canonical_smiles
>>> ('CCNC(=O)N1CCC(CC1)Nc2ncc(Cl)c(n2)c3c[nH]c4ccccc34',
     'CCNC(=O)N1CCC(C1)Nc2ncc(Cl)c(n2)c3c[nH]c4ccccc34')

# PDB structures linked to this assay
cliff.Assay.Structures
>>> [<Structure(2P33)>]

credoscript-chembl-example.py

from credoscript.extensions import chembl

pa = chembl.ProductAdaptor()

products = pa.fetch_all_by_trade_name('GLEEVEC')

for product in products:
    for f in product.Formulations:
        print f.Molecule, f.Molecule.pref_name
       
>>> <Molecule(485100)> IMATINIB MESYLATE
    <Molecule(485100)> IMATINIB MESYLATE
    <Molecule(485100)> IMATINIB MESYLATE
    <Molecule(485100)> IMATINIB MESYLATE

m = products[0].Formulations[0].Molecule

m.Hierarchy.Active
>>> <Molecule(88797)>

a = m.Hierarchy.Active

# ligands in CREDO
a.Ligands 
>>> [<Ligand(A 3 STI)>,
     <Ligand(A 1 STI)>,
     <Ligand(A 600 STI)>,
     <Ligand(B 600 STI)>,
     <Ligand(C 600 STI)>,
     <Ligand(D 600 STI)>,
     <Ligand(A 233 STI)>,
     <Ligand(B 233 STI)>]

credoscript-chembl-rdkit.py

from credoscript import *
from credoscript.extensions import chembl

ca = ChemCompAdaptor()
j07 = ca.fetch_by_het_id('J07')

ma = chembl.MoleculeAdaptor()

# fetch compounds from ChEMBL using circular fingerprints and tanimoto similarity
ma.fetch_all_by_sim(j07.ism, fp='circular', threshold=0.8)
>>> [(<Molecule(410891)>, 1.0),
     (<Molecule(410898)>, 0.948275862068966),
     (<Molecule(410900)>, 0.868852459016393),
     (<Molecule(410902)>, 0.868852459016393),
     (<Molecule(410899)>, 0.85),
     (<Molecule(410892)>, 0.819672131147541),
     (<Molecule(410905)>, 0.80952380952381),
     (<Molecule(410907)>, 0.806451612903226)]

# fetch compounds by substructure
ma.fetch_all_by_substruct(j07.ism)
>>> [<Molecule(410907)>, <Molecule(410911)>, <Molecule(410891)>]