Protein kinases (PKs) have been prominent targets for modern drug discovery against many diseases. However, PKs targeted by small-molecule inhibitors develop resistance through mutations in four types of representative mutation hotspots (gatekeeper, G-loop, αC-helix, and A-loop). Dr. Kinase represents a unique prediction server designed specifically for predicting the loci of four drug-resistance (DR) hotspots for PKs. This tool enables users to perform prediction from the protein sequence(s) or proteins with mutations.
: prediction of the loci for four DR hotsots from protein tree or sequences.
: inference of mutation effects for four DR hotspots from proteins with mutations.
Important note: Each job may take ~1 minute to finish. To facilitate fast prediction, the default prediction model is the lightweight deep-learning model (using BLOSUM50 matrix for vector embedding). Users can choose to use the protein language-based model for prediction (average 2% performance improvement, but slower). You may retrive the results anytime using the Job Identifier (JID, e.g., JOB1234_1234567890). Please do keep a record of the JID.
Example input & result of Hotspot mode:
Example input: A protein sequence in fasta format can be checked here.
Example result: A quick example result page can be checked here.
Example input & result of Mutation mode:
Example input: A protein sequence in fasta format can be checked here.
Example result: A quick example result page can be checked here.