Facilitating Colorectal Cancer Treatment by Predicting Peptide-MHC Interaction Via Deep Learning

Colorectal cancer (CRC) is a prevalent and complex disease with significant global health implications.Currently, the primary treatments for colorectal cancer in clinical practice mainly involve radiation therapy and chemotherapy, while immunotherapy such as cancer vaccines is still in its early stages.Cancer vaccines work by inducing the body's immune response against cancer-related antigens, thereby inhibiting the occurrence and progression of cancer.Currently, there is still a lack of research reports on signature antigen peptides for CRC. The discovery of new antigens with therapeutic potential is crucial for the efficacy of cancer vaccines.This study focuses on utilizing the immune epitope database (IEDB) to screen 14,681 epitopes related to CRC. A predictive model based on machine learning algorithms is used to forecast the binding affinity between novel antigens in the CRC environment and MHC-I molecules, as well as their immunogenicity. The aim is to identify new antigens with high immunogenicity and strong MHC molecule binding, thereby advancing immunotherapies, including cancer vaccines.