vaccines and immunoinformatics approaches

The paper introduces IL4Pred2, a host-specific method for predicting IL-4 inducing peptides in human and mouse, achieving high AUC and MCC scores on independent datasets.

The paper profiles HLA immunopeptidomes and interactomes in microglia, identifying disease-associated epitopes and their interactions.

This paper improves epitope prediction by using phylogeny-informed transfer learning with protein language models, showing better performance than existing methods.

ANARCII is a language model that numbers antigen receptor sequences without alignment, achieving high accuracy and efficiency for diverse sequences.

This paper introduces Spectravax, a computational method to design broad-spectrum T-cell-inducing vaccines, and demonstrates its effectiveness against Betacoronaviruses.

The paper introduces PROXICLUST, a method that uses spatial clustering of interface residues to predict viral protein binding affinities across larger mutation spaces using smaller datasets.

The paper benchmarks 17 HLA-I peptide binding prediction tools using a large dataset, identifies performance differences, and highlights the importance of dataset quality and model interpretability.

The paper designs antigens that present a tumor-specific cryptic epitope in the conformation recognized by mAb806, showing successful stabilization of the epitope in vitro.

This study improves antigen specificity prediction by fine-tuning pre-trained antibody language models on SARS-CoV-2 and influenza antigens, showing better accuracy than models trained on embeddings alone.

The paper benchmarks 49 algorithms for predicting B. pertussis vaccination outcomes, finding that purpose-built models outperform literature-based ones due to better handling of nonlinearities and data preprocessing.

This study identifies cross-reactive vaccine antigens in Gram-positive ESKAPE pathogens using subtractive proteome analysis and opsonic activity testing.

The paper updates PANDORA to predict bidirectional peptide conformations for MHC class II, achieving low RMSD values for reversed-bound peptides.

The paper identifies a common network of residue-residue contacts in peptide-MHC class II interactions across different complexes and proposes criteria for predicting peptide binding.

This study develops a multi-epitope vaccine candidate against KFDV using immunoinformatic tools, showing stable interaction with TLR2-TLR6 and codon optimization for prokaryotic expression.

This study evaluates how well MHC-I ligand predictions can be extrapolated across alleles and species using an improved predictor, finding high accuracy in human and mouse but lower in other species.

The paper presents a method using phage-displayed oligo pools and Nanopore sequencing to profile linear epitopes in a vaccine, revealing unexpected immunogenicity of epitope tags.

This paper introduces a deep learning model called GRIP that predicts T-cell receptor epitope peptides, achieving high accuracy and generating novel sequences for cancer immunotherapy.

VASCO provides a curated dataset of viral antigen-antibody complexes to improve predictive modeling and immunogen design for antiviral therapeutics.

The paper evaluates TCR structure prediction models using a large dataset, finds comparable structural diversity in alpha and beta chains, and provides over 1.5M predicted structures for further analysis.

This study evaluates AlphaFold3 and AlphaFold2-Multimer for nanobody epitope prediction, finding overall success rates below 50% and highlighting CDR3 characteristics as key factors affecting accuracy.

GRATCR generates epitope-specific TCR sequences with fewer training data than state-of-the-art models, showing higher specificity and biological functionality.

ParaSurf improves paratope prediction by integrating surface geometry and non-geometric factors, achieving state-of-the-art results across multiple metrics.

This study identifies epitopes from Rickettsia and Dengue virus proteins using reverse vaccinology and evaluates their reactivity in serological tests for vector-borne diseases.

This study enhances T cell receptor generation for novel epitopes using in-context learning and self-contemplation prompting, achieving performance comparable to or better than ground-truth TCRs.

This study shows that feature selection improves peptide binding predictions for TCRs by optimizing model complexity and performance using physicochemical properties and machine learning.

EpitopeMiner identifies immunogenic epitopes using a domain-specific LLM with RAG, improving retrieval of relevant literature and supporting neoantigen validation.

IFNBoost is a machine learning model that predicts IFNγ-inducing peptides with high accuracy, outperforming existing models and using both peptide sequences and metadata features.

SwiftMHC rapidly predicts peptide-MHC binding and generates 3D structures, outperforming sequence-based methods in speed while maintaining accuracy.

EpitopeGen generates T-cell epitopes from TCR sequences using a transformer model, outperforming baselines and identifying antigen-specific T cells in cancer and COVID-19.

This paper introduces ConBoTNet, a deep learning model that improves MHC-II peptide binding affinity prediction using supervised contrastive learning and bottleneck transformers, achieving better performance than existing methods.