vaccines and immunoinformatics approaches

The paper determines TAP structures with peptides to show how it selects peptides by length while allowing sequence diversity through terminal interactions.

This paper introduces TCR-VALID, a variational autoencoder for generating and analyzing TCR sequences, showing improved clustering and representation quality.

This study evaluates whether domain-specific protein language models outperform general models in immunology tasks, finding nuanced performance differences and highlighting key hyperparameters.

VitTCR predicts TCR-peptide interactions using a vision transformer-based model, outperforming existing methods in classification accuracy.

RACER-m predicts T cell specificity using structural features, requiring fewer training examples than sequence-based methods.

The paper fine-tunes AlphaFold using high-resolution MHC-peptide crystal structures to improve prediction accuracy for MHC-peptide complex structures, outperforming existing methods in RMSD and lDDT scores.

BacScan identifies highly conserved immunogenic bacterial proteins across serotypes, showing potential for vaccine development.

The paper develops a machine learning method to modify protein immune-visibility to CTLs, showing it can alter HIV Nef's visibility while preserving sequence diversity.

This paper introduces MixTCRpred, a deep learning model that predicts TCR-epitope interactions and identifies epitope-specific T cells, showing enrichment in COVID-19 patients.

The paper identifies a concise set of 137 virus-specific CD8 T cell epitopes recognized by human T cells, with 84% showing prevalent reactivity and associations with effector and memory phenotypes.

GraphMHC predicts neoantigen binding using graph neural networks on molecular structures, achieving high accuracy and showing survival differences in melanoma patients.

The paper identifies SARS-CoV-2 BA.1 epitopes recognized by mouse CD4+ T cells and generates corresponding tetramer reagents for studying immune responses.

X-ray structures of anti-MHC-I/MHC-I complexes define epitopes overlooked by computational prediction.

This paper uses machine learning to design cytotoxic T cell epitopes by modeling proteasomal degradation scores and testing their immunogenicity.

The paper introduces MHCXAI, an explainable AI framework to interpret deep learning predictors of MHC class I peptide presentation, validating explanations against ground truth and robustness.

NetTCR 2.2 improves TCR specificity predictions by combining pan- and peptide-specific training, loss-scaling, and sequence similarity integration, achieving state-of-the-art performance on the IMMREP 2022 benchmark.

The paper introduces two methods to map mimotopes to epitopes, including an ensemble approach and a novel algorithm called MimoTree that handles gaps in sequences.

The paper develops Dolphyn, a machine learning method to improve PhIP-Seq by prioritizing epitopes, increasing antibody-bound peptides from 10% to 31% and reducing synthesized peptides by 78%.

PoxiPred predicts potential antigens and T-cell epitopes for poxviruses using an AI-based tool trained on viral proteomes.

T-FINDER identifies T cell epitopes and their cognate receptors by analyzing HLA-presented peptides, demonstrating its utility in multiple diseases.

This study curates a dataset of influenza HA antibodies and develops an explainable language model to predict antibody specificity, identifying and validating HA stem antibodies.

This study compares clustering models for predicting T cell receptor (TCR) antigen specificity, finding variability in performance and scalability but high similarity in clusters for TCRs recognizing the same epitope.

The paper develops a hybrid model for predicting both linear and conformational B-cell epitopes using machine learning and BLAST, achieving an AUC of 0.83 on an independent dataset.

The paper introduces a database of TCR-peptide mutational assays and a Bayesian model to predict TCR cross-reactivity, showing improved performance over existing methods.

The paper examines bias in NetMHCpan training datasets and evaluates prediction accuracy on novel alleles, finding no significant impact of training data disparities on prediction quality.

HLA-EpiCheck predicts B-cell epitopes on HLA proteins using molecular dynamics simulations and outperforms existing methods, aiding in organ transplant matching.

This paper describes using MixMHC2pred to predict MHC-II binding specificity and ligands from amino acid sequences across species.

epiTCR-KDA improves TCR-peptide binding prediction by incorporating structural dihedral angles and knowledge distillation, achieving high AUC for both seen and unseen data.

The paper identifies SARS-CoV-2 peptides presented by HLA-II complexes, showing their association with CD4+ T cell epitopes and distinct targeting of viral proteins compared to HLA-I.