Dr. Kinase: predicting the drug-resistance hotspots of protein kinases.

Protein kinases (PKs) regulate various cellular functions, and are targeted by small-molecule kinase inhibitors (KIs) in cancers and other diseases. However, drug resistance (DR) of KIs occurs through critical mutations in four types of representative hotspots, including gatekeeper, G-loop, αC-helix, and A-loop. KI DR has become a common clinical complication affecting multiple cancers, targeted kinases, and drugs. To tackle this challenge, we report an upgraded web server, namely Dr. Kinase, for predicting the loci of four DR hotspots and assessing effects of mutations on DR hotspots for PKs in our previous studies, by utilizing multimodal features and deep hybrid learning. The performance of Dr. Kinase has been rigorously evaluated using independent testing, demonstrating excellent accuracy with area under the curve values exceeding 0.89 in different types of DR hotspot predictions. We further conducted in silico analyses to evaluate and validate the epidermal growth factor receptor mutations on protein conformation and KIs' binding efficacy. Dr. Kinase is freely available at http://modinfor.com/drkinase, with comprehensive annotations and visualizations. We anticipate that Dr. Kinase will be a highly useful service for the basic, translational, and clinical community to unveil the molecular mechanisms of DR and the development of next-generation KIs for emerging cancer precision medicine.