Method and system for automated screening and optimization of bispecific immunoreceptors

Abstract

The application provides a method and system for screening and optimizing a bispecific immunoreceptor, and belongs to the field of immunotherapy. The sequence of the bispecific immunoreceptor to be screened, the sequences of the first and second target antigens, and the mutation domain of the bispecific immunoreceptor are obtained; the sequence of the bispecific immunoreceptor and the two target antigens are subjected to parallel conformation prediction based on a structure prediction model, and the optimal conformation is screened based on a confidence level; based on the mutation domain of the bispecific immunoreceptor in the optimal conformation, a sequence generation model is used to generate candidate objects carrying different mutation combinations; a multi-objective scoring function is constructed, and the candidate objects are subjected to at least two rounds of screening: in the first round, the first candidate set is screened according to stability and interface indexes, and in the second round, parallel docking simulation calculation is performed on the first candidate set and the two target antigens to obtain docking energy and interface residue annotation, the scoring function is updated based on the calculation results to obtain an optimized bispecific immunoreceptor, and the automatic screening and optimization of the bispecific immunoreceptor are realized.

Images