SimBoost: a read-across approach for predicting drug-target binding affinities using gradient boosting machines

摘要: Computational prediction of the interaction between drugs and targets is a standing challenge in field drug discovery. A number rather accurate predictions were reported for various binary drug–target benchmark datasets. However, notable drawback representation data that missing endpoints non-interacting pairs are not differentiated from inactive cases, predicted levels activity depend on pre-defined binarization thresholds. In this paper, we present method called SimBoost predicts continuous (non-binary) values binding affinities compounds proteins thus incorporates whole spectrum true negative to positive interactions. Additionally, propose version SimBoostQuant which computes interval order assess confidence affinity, defining Applicability Domain metrics explicitly. We evaluate two established datasets one new dataset use as read-across cheminformatics applications. demonstrate our methods outperform previously models across studied