Fig. 2From: Exploring the ability of machine learning-based virtual screening models to identify the functional groups responsible for bindingNon-linear functions used to determine the interaction score of a synthetic residue-ligand atom interaction; both functions are proportional to the Probability Density Function of a Gamma (4, 1) distribution. Separate functions are used for Polar interactions and Hydrophobic interactions, making it more difficult for the model to learn the deterministic binding ruleBack to article page