Basic Usage Tutorial¶

If you want to rapidly translate an internal library of SMILES structures into the CDI embedding maps, utilizing the basic Python API is easiest.

Ensure you've installed ChemicalDice (Tier 1).

1. Local Initialization¶

import pandas as pd
from ChemicalDice.core.api_client import collect_features_from_csv

2. Setting Up Data¶

Your CSV requires exactly one column titled SMILES. ChemicalDice uses this structure strictly.

SMILES,Assay_Target,Assay_Result
CC(=O)Oc1ccccc1C(=O)O,Aspirin,Active

Any other columns (like Assay_Target) will be ignored during the actual extraction iteration securely.

3. Invoking the Extractor¶

df_results = collect_features_from_csv(
    filepath="compounds.csv",
    convert_to_canonical=True # Enables safety RDKit filtering
)

print(df_results.head())
# Output:
# SMILES                    CDI1      CDI2      CDI3  ... CDI8192
# CC(=O)Oc1ccccc1C(=O)O    0.4121   -1.1092    0.052 ....