Activark is a data-driven, ML-based approach to predict the functional consequence of genetic changes in protein kinases. Activark was trained on a curated dataset of activating (i.e. constitutive-activation or increase in kinase activity), deactivating (i.e. loss or decrease in Kinase activity), and drug-resistance protein variants in human kinases and using sequence and structural features. The following sections describe the methodology used to develop Activark.
To jump to a specific section on this page, click on the links below: