smiles = df_descriptors.ix[id, 'Structure']
descriptors = list(df_descriptors.ix[id])[2:]
normalized = [(f - min_bounds[i]) / (max_bounds[i] - min_bounds[i])
for i, f in enumerate(descriptors)]
fp = pack_normalized_descriptors_to_fingerprint(
normalized, density=0.3, byte_size=fp_size_bytes)
mol = indigo.loadMolecule(smiles)
molecules += [mol]
ext_fp = mol.fingerprintExt(fp, fp_size_bytes)
bingo.insertWithExtFP(mol, ext_fp)
print("Similarity matrix:")
for m1 in molecules:
for m2 in molecules:
similarity = indigo.similarity(m1, m2)
print("%.4f " % similarity, end="")
print()
print("Bingo similarity search:")
query_id = 0 # use the fist molecule as a query for an example
smiles = df_descriptors.ix[query_id, 'Structure']
print("Query: %s" % smiles)
descriptors = list(df_descriptors.ix[query_id])[2:]
normalized = [(f - min_bounds[i]) / (max_bounds[i] - min_bounds[i])
for i, f in enumerate(descriptors)]
fp = pack_normalized_descriptors_to_fingerprint(normalized,
density=0.3,
byte_size=fp_size_bytes)
mol = indigo.loadMolecule(smiles)
df_descriptors = []
indigo = Indigo()
# indigo.setOption("similarity-type", "sim")
# indigo.setOption("similarity-type", "chem")
# indigo.setOption("similarity-type", "ECFP2")
indigo.setOption("similarity-type", "ECFP4")
# indigo.setOption("similarity-type", "ECFP6")
# indigo.setOption("similarity-type", "ECFP8")
fingerprints = []
for sml in smiles:
mol = indigo.loadMolecule(sml)
fp = mol.fingerprint("sim")
fingerprints += [fp]
print("*** Fingerprints ***")
for i, fp in enumerate(fingerprints):
str = fp.toString()
density = fp_density(str)
print("#%2d: Density: %f ; FP: %s" % (i, density, str))
print("*** Similarity matrix ***")
for f0 in fingerprints:
for f1 in fingerprints:
similarity = indigo.similarity(f0, f1)
print(" %.3f" % similarity, end="")
print("\n")
