From python API

Transform molecules into fingerprints

Create an object of a specific fingerprint class. Then use the fit method to load the sdf file, and transform to obtain the fingerprints.

from drug_learning.two_dimensions.Input import fingerprints as fp

input_sdf = "../datasets/ligands.sdf"

mor = fp.MorganFP()
structures = mor.fit(input_sdf)
features = mor.transform()

Set the applicability domain

from drug_learning.two_dimensions.Input import applicability_domain as ad

AD = ad.ApplicabilityDomain()
AD.fit(train_features)
AD.predict(test_features)