Procedure used to build the EGFR Mutant Structural Database. First, we applied Rosetta ddg_monomer protocol and comparative modeling (CM) protocol to predict EGFR mutant structures. Secondly, the predicted structures were refined with a minimization step using sander in Amber. Then a drug (gefitinib or erlotinib) was added to the mutant structures followed by MD simulation. Subsequently, we employed MM-GBSA in Amber to calculate the binding free energies of the EGFR mutants and the inhibitor. Finally, the refined mutant structures and their corresponding binding free energies with gefitinib and erlotinib were collected to establish the database.