MHC-I epitope affinity prediction method based on deep learning

Abstract

The invention discloses a method for predicting MHC-I epitope affinity based on deep learning, which includes: obtaining a plurality of polypeptides through a public database; converting the polypeptides into 21mer peptides according to the binding mode of MHC-I molecules and peptides; extracting the peptides The characteristics of the polypeptide, the characteristics include: sequence characteristics, hydrophilic characteristics, polar characteristics and position characteristics; respectively carry out characteristic encoding on the characteristics of the polypeptide to obtain a 4*21-dimensional characteristic matrix; the public database The polypeptide data in is used as a training set for model training, and according to the classification of the alleles of the polypeptide, the feature matrix of the polypeptide is input into the pre-established CNN model to establish a prediction model, and the number of the established prediction models is the same as that of the polypeptide Corresponding to the classification data of the alleles of ; the binding affinity test was performed using the polypeptide data of the public database as the validation set of the prediction model. The application can effectively predict the MHC-I epitope affinity, and the prediction accuracy is higher and more stable.

Description

technical field [0001] The present invention relates to the fields of biological information and tumor immunotherapy, in particular to a method for predicting the affinity of MHC-I epitopes based on deep learning. Background technique [0002] Since neoantigens are ideal targets for immunotherapy, understanding the binding affinity between specific peptides and alleles of MHC is an essential step in designing vaccines. The large number of peptide chains makes research time-consuming and labor-intensive. With advances in sequencing technology and bioinformatics, predicting binding affinities between peptides and MHC alleles has become more flexible and economical. [0003] The MHC (Major Histocompatibility Complex) is a family of genes found in most vertebrate genomes and is closely related to the immune system. Human MHC is also known as human leukocyte antigen (HLA). There are two types of MHC. The first MHC (MHC-I) deals with the internal breakdown of proteins such as ...

AI Technical Summary

Problems solved by technology

However, technical solutions to accurately predict the binding affinity between MHC-I proteins and their peptides are currently lacking

Images