Second enumerate example returns different output

[olivertomic]

In the second example the second line has one command too many. >>> peptide_backbone = PeptideBuilder(2) should be on the next line.

When running this example I get different ouputs.

from cocktail_shaker import Cocktail
from cocktail_shaker import PeptideBuilder
peptide_backbone = PeptideBuilder(2)
cocktail = Cocktail(peptide_backbone,ligand_library = ['Br', 'I'])
combinations = cocktail.shake()
print (combinations)
['NC(Br)C(=O)NC(I)C(=O)NCC(=O)O', 'NC(I)C(=O)NC(Br)C(=O)NCC(=O)O']
enumerations = cocktail.enumerate(enumeration_complexity='low')
print (len(enumerations))
20
enumerations = cocktail.enumerate(enumeration_complexity='med')
print (len(enumerations))
186
enumerations = cocktail.enumerate(enumeration_complexity='high')
print (len(enumerations))
1789

My output is the following (note that the numerical outputs are different):

Generating 2 Compounds...
['NC(Br)C(=O)NC(I)C(=O)NCC(=O)O', 'NC(I)C(=O)NC(Br)C(=O)NCC(=O)O']
Enumerating 2 Compounds....
192
Enumerating 2 Compounds....
20
Enumerating 2 Compounds....
1792

[Sulstice]

Ah I need to note something in the documentation here. What ends up happening and this is somewhat of a bug of the algorithm I implemented not so much the code and its pretty glaring. So the enumeration happens something like this

molecule = Chem.MolFromSmiles(self.combinations[i])
smiles_enumerated = Chem.MolToSmiles(molecule, doRandom=True)
if dimensionality == '1D' and smiles_enumerated not in enumerated_molecules:
    enumerated_molecules.append(smiles_enumerated)

So the algorithm itself is inconsistent because we are generating random representations of the string and then seeing if it previously existed in our unique list. The amount of times we try depends on the user sending in different complexity values.

I'm going to label this a bug and something to enhance in future releases.

[Sulstice]

I'll also include this in the documentation that your results may vary compared to the example.