Investigation of mutations to customize structurally dynamic papain proteins for temperature-specific peptide binding by complementary use of two different artificial intelligence methods and molecular simulations

Customization of proteins to undertake temperature-specific functions would expand their scope of use in medical treatment, food processing, and bioelectronic devices. To customize papain for temperature-specific peptide binding, the dynamic structure of papain was modified by repeatedly mutating V110, R111, and Q112 through the complementary use of deep neural network methods and molecular dynamics simulations. Overall, 18 mutation patterns were discovered to customize papain for temperature-specific peptide binding. The decision tree indicated that the binding energy changed rapidly, with the 112th residue and carbon atoms of the 110 to 112th residue as the main influencing factors.