Artificial intelligence applied in neoantigen identification facilitates personalized cancer immunotherapy

Abstract

The field of cancer neoantigen investigation has developed swiftly in the past decade. Predicting novel and true neoantigens derived from large multi-omics data became difficult but critical challenges. The rise of Artificial Intelligence (AI) or Machine Learning (ML) in biomedicine application has brought benefits to strengthen the current computational pipeline for neoantigen prediction. ML algorithms offer powerful tools to recognize the multidimensional nature of the omics data and therefore extract the key neoantigen features enabling a successful discovery of new neoantigens. The present review aims to outline the significant technology progress of machine learning approaches, especially the newly deep learning tools and pipelines, that were recently applied in neoantigen prediction. In this review article, we summarize the current state-of-the-art tools developed to predict neoantigens. The standard workflow includes calling genetic variants in paired tumor and blood samples, and rating the binding affinity between mutated peptide, MHC (I and II) and T cell receptor (TCR), followed by characterizing the immunogenicity of tumor epitopes. More specifically, we highlight the outstanding feature extraction tools and multi-layer neural network architectures in typical ML models. It is noted that more integrated neoantigen-predicting pipelines are constructed with hybrid or combined ML algorithms instead of conventional machine learning models. In addition, the trends and challenges in further optimizing and integrating the existing pipelines are discussed.

Artificial intelligence and neoantigens: paving the path for precision cancer immunotherapy

“…For best practices in variant calling in clinical sequencing, readers are referred to the work of Koboldt (33). A comprehensive overview of the variant calling tools and their pros and cons is provided in the paper of Cai et al (25).…”

Improving generalizability for MHC-I binding peptide predictions through geometric deep learning

“…Description SkipGram [6,1], float Skip-Gram embedding for residue type Edesolv [1], float Desolvation energy, calculated as in Dominguez et al 67 Anchor [1], bool Label to indicate if the residue is an anchor residue (peptide side) or a pocket residue (MHC side) RCD_apolar-apolar [1], float Residue Contact Density for apolar-apolar interactions calculated as in Vangone and Bonvin 69 RCD_apolar-charged [1]…”

“…The accurate identification of peptides presented by MHC on the cell surface is crucial for understanding autoimmune diseases 1 , recognizing pathogens 2 , and addressing transplant rejection 3 . The recent notable clinical advancements in cancer immunotherapies 4,5 , specifically targeting tumor-associated or tumor-specific peptide-MHC complexes, underscore the urgency to advance computational methods for identifying MHC-bound peptides 6,7 . With over 14,000 human MHC-I proteins 8 encoded by the canonical HLA genes (HLA-A, HLA-B, and HLA-C), the theoretical 9-residue peptides (called 20 9 9-mers) create impractical in vitro testing scenarios.…”

Combination of multiple omics techniques for a personalized therapy or treatment selection

“…In recent years, next to “standard” HLA class I-restricted peptides, these strategies have identified many neoAg-restricted to HLA class II molecules ( 20 , 21 ) or derived from non-coding sequences ( 22 ) leading to the development of new algorithms for their improved identification from sequencing data ( 23 – 25 ). Continuous progresses in artificial intelligence approaches are further improving the capabilities to identify neoAg for clinical application ( 26 ).…”