Awesome

enamine-real-search

⚠️ This no longer works! Please go to https://github.com/matteoferla/Python_SmallWorld_API for a newer, working alternative to this.

Wrapper around similarity and substructure searching functionality on enaminestore.com

Installation

pip install ers

Usage

To do a similarity search:

from search import EnamineSession

session = EnamineSession()
similarity_results = session.similarity_search(smiles='C1=CC=C(C=C1)O', threshold=0.1)

To do a substructure search:

from search import EnamineSession

session = EnamineSession()
substructure_results = session.substructure_search(smiles='C1=CC=C(C=C1)O')

Output

For both types of search, a list of dictionaries our output, with one dictionary in the list for each molecule that is a match from the search. It's super easy to turn these into a dataframe with pandas and view the results as a table (there's a full example in example.ipynb.

The keys in the molecule dictionaries and their meanings are:

• vendorId - the vendor ID that enamine uses (use this if asking for a quote from Enamine)
• smiles - the smiles string of the matching molecule
• fsp3 - the fraction of sp3 carbon atoms
• hac - the heavy atom count
• hba - the number of hydrogen-bond acceptors
• hbd - the number of hydrogen bond donors
• logP - the log of the partition coefficient between octanol and water (hydrophobicity)
• mw - molecular weight
• rotb - the number of rotatable bonds
• tpsa - topological polar surface area
• priceCategory - S=simple, M=Advanced (check with Enamine)

And for the similarity search, there is an additional field for similarity, where a value of to 1 indicates the molecule is an exact match to the input molecule (probably not inc. tautomers etc.) and 0 is not a match at all.