Journal: Journal of chemical information and modeling
Article Title: Predictive models for human cytochrome P450 3A7 selective inhibitors and substrates
doi: 10.1021/acs.jcim.2c01516
Figure Lengend Snippet: Parameter sets for each optimal model. (A) Feature selection methods and data rebalancing approaches. (B) Number of ECFP4 features that generated the optimal model for each CYP3A4 inhibitor/substrate dataset. These parameter combinations were applied as a preprocessing step prior to modeling.
Article Snippet: The ECFP4 fingerprints (1024 bits) were generated using the Chemistry Development Kit (CDK) package within the Konstanz Information Miner open source software (KNIME, version 4.0.2, https://www.knime.org/ ) 20 .
Techniques: Selection, Generated