How to use the selfies.encoder function in selfies
To help you get started, we’ve selected a few selfies examples, based on popular ways it is used in public projects.
Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
aspuru-guzik-group / selfies / VariationalAutoEncoder_with_SELFIES / chemistryVAE.py View on Github 
- selfies alphabet
- longest selfies string
- smiles encoding (equivalent to file content)
- smiles alphabet (character based)
- longest smiles string
"""
df = pd.read_csv(filename_data_set_file_smiles)
smiles_list = np.asanyarray(df.smiles)
smiles_alphabet=list(set(''.join(smiles_list)))
largest_smiles_len=len(max(smiles_list, key=len))
selfies_list=[]
selfies_len=[]
print('--> Translating SMILES to SELFIES...')
for individual_smile in smiles_list:
individual_selfie=selfies.encoder(individual_smile)
selfies_list.append(individual_selfie)
selfies_len.append(len(individual_selfie)-len(individual_selfie.replace('[',''))) # len of SELFIES
selfies_alphabet_pre=list(set(''.join(selfies_list)[1:-1].split('][')))
selfies_alphabet=[]
for selfies_element in selfies_alphabet_pre:
selfies_alphabet.append('['+selfies_element+']')
largest_selfies_len=max(selfies_len)
print('Finished translating SMILES to SELFIES.')
return(selfies_list, selfies_alphabet, largest_selfies_len, smiles_list, smiles_alphabet, largest_smiles_len)
def mol2string(mol):
Chem.Kekulize(mol, clearAromaticFlags=True)
smiles = Chem.MolToSmiles(mol, canonical=False)
if string_type == 'selfies':
return encoder(smiles).split('][')
if string_type == 'deepsmiles':
string = converter.encode(smiles)
return list(string)
return list(smiles)print('test_molecule1: '+test_molecule1+'\n')
print('selfies1: '+selfies1+'\n')
print('smiles1: '+smiles1+'\n')
print('equal: '+str(test_molecule1==smiles1)+'\n\n\n')
test_molecule2='CC(C)c1noc(-c2cc[nH+]c(N3CCN(C(=O)[C@H]4C[C@H]4C)CC3)c2)n1' # from ZINC database
selfies2=encoder(test_molecule2)
smiles2=decoder(selfies2)
print('test_molecule2: '+test_molecule2+'\n')
print('selfies2: '+selfies2+'\n')
print('smiles2: '+smiles2+'\n')
print('equal: '+str(test_molecule2==smiles2)+'\n\n\n')
test_molecule3='CCOC(=O)C1(C(=O)OCC)C23c4c5c6c7c8c4-c4c2c2c9c%10c4C4%11c%12c-%10c%10c%13c%14c%15c%16c%17c%18c%19c%20c%21c%22c%23c%24c(c-7c(c7c%12c%13c(c7%24)c(c%19%23)c%18%14)C84C%11(C(=O)OCC)C(=O)OCC)C%224C(C(=O)OCC)(C(=O)OCC)C64c4c-5c5c6c(c4-%21)C%204C(C(=O)OCC)(C(=O)OCC)C%174c4c-6c(c-2c(c4-%16)C92C(C(=O)OCC)(C(=O)OCC)C%10%152)C513' # from PubChem
selfies3=encoder(test_molecule3)
smiles3=decoder(selfies3)
print('test_molecule1: '+test_molecule3+'\n')
print('selfies1: '+selfies3+'\n')
print('smiles1: '+smiles3+'\n')
print('equal: '+str(test_molecule3==smiles3)+'\n\n\n')
#Create a random Molecule
my_alphabet=['[Branch1_1]','[Branch1_2]','[Branch1_3]','[Ring1]','[O]','[=O]','[N]','[=N]','[C]','[=C]','[#C]','[S]','[=S]','[P]','[F]']; # this is a very small alphabet from which the random selfies are generated
# This alphabet can be extended with additional elements. For example, see the list start_alphabet in the function smiles_to_selfies.
# Also when you run the three test-molecules above, you see the brackets that are used, and can use some of them.
len_of_molecule=30 # Number of selfies symbols of the random string. The final SMILES string will not necessarily be of the same size, because some elements of this alphabet stop the derivation (such as Flour, as it can form only a single bond)
rnd_selfies=''
for ii in range(len_of_molecule):print('test_molecule1: '+test_molecule1+'\n')
print('selfies1: '+selfies1+'\n')
print('smiles1: '+smiles1+'\n')
print('equal: '+str(test_molecule1==smiles1)+'\n\n\n')
test_molecule2='CC(C)c1noc(-c2cc[nH+]c(N3CCN(C(=O)[C@H]4C[C@H]4C)CC3)c2)n1' # from ZINC database
selfies2=encoder(test_molecule2)
smiles2=decoder(selfies2)
print('test_molecule2: '+test_molecule2+'\n')
print('selfies2: '+selfies2+'\n')
print('smiles2: '+smiles2+'\n')
print('equal: '+str(test_molecule2==smiles2)+'\n\n\n')
test_molecule3='CCOC(=O)C1(C(=O)OCC)C23c4c5c6c7c8c4-c4c2c2c9c%10c4C4%11c%12c-%10c%10c%13c%14c%15c%16c%17c%18c%19c%20c%21c%22c%23c%24c(c-7c(c7c%12c%13c(c7%24)c(c%19%23)c%18%14)C84C%11(C(=O)OCC)C(=O)OCC)C%224C(C(=O)OCC)(C(=O)OCC)C64c4c-5c5c6c(c4-%21)C%204C(C(=O)OCC)(C(=O)OCC)C%174c4c-6c(c-2c(c4-%16)C92C(C(=O)OCC)(C(=O)OCC)C%10%152)C513' # from PubChem
selfies3=encoder(test_molecule3)
smiles3=decoder(selfies3)
print('test_molecule3: '+test_molecule3+'\n')
print('selfies3: '+selfies3+'\n')
print('smiles3: '+smiles3+'\n')
print('equal: '+str(test_molecule3==smiles3)+'\n\n\n')
test_molecule4='Cc1c(C)c(S(=O)(=O)NC(=N)NCCC[C@H](NC(=O)[C@@H]2CCCN2C(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](C)NC(=O)[C@@H]2CCCN2C(=O)[C@@H]2CCCN2C(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](COC(C)(C)C)NC(=O)[C@H](CCC(=O)OC(C)(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](CCCNC(=N)NS(=O)(=O)c2c(C)c(C)c3c(c2C)CCC(C)(C)O3)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@@H](NC(=O)[C@H](CCCNC(=N)NS(=O)(=O)c2c(C)c(C)c3c(c2C)CCC(C)(C)O3)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](Cc2cn(C(=O)OC(C)(C)C)cn2)NC(=O)[C@H](CCC(=O)OC(C)(C)C)NC(=O)[C@@H]2CCCN2C(=O)[C@H](COC(C)(C)C)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](Cc2ccccc2)NC(=O)[C@H](COC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](COC(C)(C)C)NC(=O)CNC(=O)OC(C)(C)C)C(C)C)C(=O)O)c(C)c2c1OC(C)(C)CC2'
selfies4=encoder(test_molecule4)
smiles4=decoder(selfies4)
print('test_molecule4: '+test_molecule4+'\n')
print('selfies4: '+selfies4+'\n')
print('smiles4: '+smiles4+'\n')
print('equal: '+str(test_molecule4==smiles4)+'\n\n\n')print('test_molecule2: '+test_molecule2+'\n')
print('selfies2: '+selfies2+'\n')
print('smiles2: '+smiles2+'\n')
print('equal: '+str(test_molecule2==smiles2)+'\n\n\n')
test_molecule3='CCOC(=O)C1(C(=O)OCC)C23c4c5c6c7c8c4-c4c2c2c9c%10c4C4%11c%12c-%10c%10c%13c%14c%15c%16c%17c%18c%19c%20c%21c%22c%23c%24c(c-7c(c7c%12c%13c(c7%24)c(c%19%23)c%18%14)C84C%11(C(=O)OCC)C(=O)OCC)C%224C(C(=O)OCC)(C(=O)OCC)C64c4c-5c5c6c(c4-%21)C%204C(C(=O)OCC)(C(=O)OCC)C%174c4c-6c(c-2c(c4-%16)C92C(C(=O)OCC)(C(=O)OCC)C%10%152)C513' # from PubChem
selfies3=encoder(test_molecule3)
smiles3=decoder(selfies3)
print('test_molecule3: '+test_molecule3+'\n')
print('selfies3: '+selfies3+'\n')
print('smiles3: '+smiles3+'\n')
print('equal: '+str(test_molecule3==smiles3)+'\n\n\n')
test_molecule4='Cc1c(C)c(S(=O)(=O)NC(=N)NCCC[C@H](NC(=O)[C@@H]2CCCN2C(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](C)NC(=O)[C@@H]2CCCN2C(=O)[C@@H]2CCCN2C(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](COC(C)(C)C)NC(=O)[C@H](CCC(=O)OC(C)(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](CCCNC(=N)NS(=O)(=O)c2c(C)c(C)c3c(c2C)CCC(C)(C)O3)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@@H](NC(=O)[C@H](CCCNC(=N)NS(=O)(=O)c2c(C)c(C)c3c(c2C)CCC(C)(C)O3)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](Cc2cn(C(=O)OC(C)(C)C)cn2)NC(=O)[C@H](CCC(=O)OC(C)(C)C)NC(=O)[C@@H]2CCCN2C(=O)[C@H](COC(C)(C)C)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](Cc2ccccc2)NC(=O)[C@H](COC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](COC(C)(C)C)NC(=O)CNC(=O)OC(C)(C)C)C(C)C)C(=O)O)c(C)c2c1OC(C)(C)CC2'
selfies4=encoder(test_molecule4)
smiles4=decoder(selfies4)
print('test_molecule4: '+test_molecule4+'\n')
print('selfies4: '+selfies4+'\n')
print('smiles4: '+smiles4+'\n')
print('equal: '+str(test_molecule4==smiles4)+'\n\n\n')
#Create a random Molecule, test robustness
my_alphabet=selfies_alphabet() # this is a very small alphabet from which the random selfies are generated
# This alphabet can be extended with additional elements. For example, see the list start_alphabet in the function smiles_to_selfies.
# Also when you run the three test-molecules above, you see the brackets that are used, and can use some of them.#
#
# For comments, bug reports or feature ideas, please send an email to
# mario.krenn@utoronto.ca and alan@aspuru.com
# =============================================================================
from random import randint
from selfies import encoder, decoder, selfies_alphabet
# Now we encode three molecules from SMILES -> SELFIES, and decode them from SELFIES -> SMILES
test_molecule1='CN1C(=O)C2=C(c3cc4c(s3)-c3sc(-c5ncc(C#N)s5)cc3C43OCCO3)N(C)C(=O)C2=C1c1cc2c(s1)-c1sc(-c3ncc(C#N)s3)cc1C21OCCO1' # non-fullerene acceptors for organic solar cells
selfies1=encoder(test_molecule1)
smiles1=decoder(selfies1)
print('test_molecule1: '+test_molecule1+'\n')
print('selfies1: '+selfies1+'\n')
print('smiles1: '+smiles1+'\n')
print('equal: '+str(test_molecule1==smiles1)+'\n\n\n')
test_molecule2='CC(C)c1noc(-c2cc[nH+]c(N3CCN(C(=O)[C@H]4C[C@H]4C)CC3)c2)n1' # from ZINC database
selfies2=encoder(test_molecule2)
smiles2=decoder(selfies2)
print('test_molecule2: '+test_molecule2+'\n')
print('selfies2: '+selfies2+'\n')
print('smiles2: '+smiles2+'\n')
print('equal: '+str(test_molecule2==smiles2)+'\n\n\n')
test_molecule3='CCOC(=O)C1(C(=O)OCC)C23c4c5c6c7c8c4-c4c2c2c9c%10c4C4%11c%12c-%10c%10c%13c%14c%15c%16c%17c%18c%19c%20c%21c%22c%23c%24c(c-7c(c7c%12c%13c(c7%24)c(c%19%23)c%18%14)C84C%11(C(=O)OCC)C(=O)OCC)C%224C(C(=O)OCC)(C(=O)OCC)C64c4c-5c5c6c(c4-%21)C%204C(C(=O)OCC)(C(=O)OCC)C%174c4c-6c(c-2c(c4-%16)C92C(C(=O)OCC)(C(=O)OCC)C%10%152)C513' # from PubChem# 0.1.1 (04.06.2019):
# - initial release
#
#
# For comments, bug reports or feature ideas, please send an email to
# mario.krenn@utoronto.ca and alan@aspuru.com
# =============================================================================
from random import randint
from selfies import encoder, decoder
print('SELFIES 0.2.0 - example file')
# Now we encode three molecules from SMILES -> SELFIES, and decode them from SELFIES -> SMILES
test_molecule1='CN1C(=O)C2=C(c3cc4c(s3)-c3sc(-c5ncc(C#N)s5)cc3C43OCCO3)N(C)C(=O)C2=C1c1cc2c(s1)-c1sc(-c3ncc(C#N)s3)cc1C21OCCO1' # non-fullerene acceptors for organic solar cells
selfies1=encoder(test_molecule1)
smiles1=decoder(selfies1)
print('test_molecule1: '+test_molecule1+'\n')
print('selfies1: '+selfies1+'\n')
print('smiles1: '+smiles1+'\n')
print('equal: '+str(test_molecule1==smiles1)+'\n\n\n')
test_molecule2='CC(C)c1noc(-c2cc[nH+]c(N3CCN(C(=O)[C@H]4C[C@H]4C)CC3)c2)n1' # from ZINC database
selfies2=encoder(test_molecule2)
smiles2=decoder(selfies2)
print('test_molecule2: '+test_molecule2+'\n')
print('selfies2: '+selfies2+'\n')
print('smiles2: '+smiles2+'\n')
print('equal: '+str(test_molecule2==smiles2)+'\n\n\n')
test_molecule3='CCOC(=O)C1(C(=O)OCC)C23c4c5c6c7c8c4-c4c2c2c9c%10c4C4%11c%12c-%10c%10c%13c%14c%15c%16c%17c%18c%19c%20c%21c%22c%23c%24c(c-7c(c7c%12c%13c(c7%24)c(c%19%23)c%18%14)C84C%11(C(=O)OCC)C(=O)OCC)C%224C(C(=O)OCC)(C(=O)OCC)C64c4c-5c5c6c(c4-%21)C%204C(C(=O)OCC)(C(=O)OCC)C%174c4c-6c(c-2c(c4-%16)C92C(C(=O)OCC)(C(=O)OCC)C%10%152)C513' # from PubChem# Now we encode three molecules from SMILES -> SELFIES, and decode them from SELFIES -> SMILES
test_molecule1='CN1C(=O)C2=C(c3cc4c(s3)-c3sc(-c5ncc(C#N)s5)cc3C43OCCO3)N(C)C(=O)C2=C1c1cc2c(s1)-c1sc(-c3ncc(C#N)s3)cc1C21OCCO1' # non-fullerene acceptors for organic solar cells
selfies1=encoder(test_molecule1)
smiles1=decoder(selfies1)
print('test_molecule1: '+test_molecule1+'\n')
print('selfies1: '+selfies1+'\n')
print('smiles1: '+smiles1+'\n')
print('equal: '+str(test_molecule1==smiles1)+'\n\n\n')
test_molecule2='CC(C)c1noc(-c2cc[nH+]c(N3CCN(C(=O)[C@H]4C[C@H]4C)CC3)c2)n1' # from ZINC database
selfies2=encoder(test_molecule2)
smiles2=decoder(selfies2)
print('test_molecule2: '+test_molecule2+'\n')
print('selfies2: '+selfies2+'\n')
print('smiles2: '+smiles2+'\n')
print('equal: '+str(test_molecule2==smiles2)+'\n\n\n')
test_molecule3='CCOC(=O)C1(C(=O)OCC)C23c4c5c6c7c8c4-c4c2c2c9c%10c4C4%11c%12c-%10c%10c%13c%14c%15c%16c%17c%18c%19c%20c%21c%22c%23c%24c(c-7c(c7c%12c%13c(c7%24)c(c%19%23)c%18%14)C84C%11(C(=O)OCC)C(=O)OCC)C%224C(C(=O)OCC)(C(=O)OCC)C64c4c-5c5c6c(c4-%21)C%204C(C(=O)OCC)(C(=O)OCC)C%174c4c-6c(c-2c(c4-%16)C92C(C(=O)OCC)(C(=O)OCC)C%10%152)C513' # from PubChem
selfies3=encoder(test_molecule3)
smiles3=decoder(selfies3)
print('test_molecule3: '+test_molecule3+'\n')
print('selfies3: '+selfies3+'\n')
print('smiles3: '+smiles3+'\n')
print('equal: '+str(test_molecule3==smiles3)+'\n\n\n')
test_molecule4='Cc1c(C)c(S(=O)(=O)NC(=N)NCCC[C@H](NC(=O)[C@@H]2CCCN2C(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](C)NC(=O)[C@@H]2CCCN2C(=O)[C@@H]2CCCN2C(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](COC(C)(C)C)NC(=O)[C@H](CCC(=O)OC(C)(C)C)NC(=O)[C@H](CCCCNC(=O)OC(C)(C)C)NC(=O)[C@H](CCCNC(=N)NS(=O)(=O)c2c(C)c(C)c3c(c2C)CCC(C)(C)O3)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@@H](NC(=O)[C@H](CCCNC(=N)NS(=O)(=O)c2c(C)c(C)c3c(c2C)CCC(C)(C)O3)NC(=O)[C@H](CCC(=O)NC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](Cc2cn(C(=O)OC(C)(C)C)cn2)NC(=O)[C@H](CCC(=O)OC(C)(C)C)NC(=O)[C@@H]2CCCN2C(=O)[C@H](COC(C)(C)C)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](Cc2ccccc2)NC(=O)[C@H](COC(c2ccccc2)(c2ccccc2)c2ccccc2)NC(=O)[C@H](COC(C)(C)C)NC(=O)CNC(=O)OC(C)(C)C)C(C)C)C(=O)O)c(C)c2c1OC(C)(C)CC2'selfies
SELFIES (SELF-referencIng Embedded Strings) is a general-purpose, sequence-based, robust representation of semantically constrained graphs.
