Kushwaha, A. | Search bioRxiv

bioRxiv2025

IMGT(R) at scale: FAIR, Dynamic and Automated Tools for Immune Locus Analysis

IMGT(R), the international ImMunoGeneTics information system(R), has advanced its comprehensive platform for the analysis of immunoglobulin (IG) and T cell receptor (TR) genes through the development of new automated and scalable tools. This article presents major updates aligned with IMGTs three axes of research. Axis I introduces dynamic resources such as IMGT/GeneTables, IMGT/AssemblyComparison, and IMGT/StatAssembly, enabling real-time access to annotated genomic data and quality assessment of assemblies. Axis II enhances repertoire analysis with a redesigned IMGT/GeneFrequency tool and new customization features in IMGT/V-QUEST, supporting flexible exploration of IG and TR gene expression. Axis III improves the accurate prediction of peptide-MHC thanks to IMGT/RobustpMHC. Additionally, the IMGT Knowledge Graph (IMGT-KG) and its therapeutic extension, IMGT/mAb-KG, provide semantically structured access to more than 100 million immunogenetic triplets, integrating IMGT databases and linking IMGT content to external biomedical resources. These developments promote standardization, interoperability, and integrative analysis across immunogenetics and clinical applications, reinforcing IMGTs role as a core reference in the era of FAIR data and personalized medicine.

bioinformatics↗

bioRxiv2023

IMGT/RobustpMHC: Robust Training for class-I MHCPeptide Binding Prediction

The accurate prediction of peptide-MHC class I binding probabilities is a critical endeavor in immunoinformatics, with broad implications for vaccine development and immunotherapies. While recent deep neural network based approaches have showcased promise in peptide-MHC prediction, they have two shortcomings: (i) they rely on hand-crafted pseudo-sequence extraction, (ii) they do not generalise well to different datasets, which limits the practicality of these approaches. In this paper, we present PerceiverpMHC that is able to learn accurate representations on full-sequences by leveraging efficient transformer based architectures. Additionally, we propose IMGT/RobustpMHC that harnesses the potential of unlabeled data in improving the robustness of peptide-MHC binding predictions through a self-supervised learning strategy. We extensively evaluate RobustpMHC on 8 different datasets and showcase the improvements over the state-of-the-art approaches. Finally, we compile CrystalIMGT, a crystallography verified dataset that presents a challenge to existing approaches due to significantly different peptide-MHC distributions.

bioinformatics↗