MS-BCR-DB: an integrated BCR repertoire database to mine humoral multiple sclerosis signatures

The MS-BCR-Database is a harmonized collection of publicly available human Multiple Sclerosis (MS) B-cell receptor (BCR) sequencing datasets, processed through a standardized bioinformatic pipeline and organized in an AIRR-compliant format. The database integrates 12 independent MS BCR-sequencing studies, comprising: 121 patients with MS 6,799,429 BCR sequences from MS samples Samples derived from multiple biological compartments, including cerebrospinal fluid (CSF) and peripheral blood Both bulk and single-cell sequencing datasets 5,641,741 BCR sequences from healthy controls, including both study-matched healthy controls, when available, and an independent reference healthy-control dataset from "Maturation of the Human Immunoglobulin Heavy Chain Repertoire With Age" The MS-BCR-DB additionally includes BCR sequences reconstructed from MS-related RNA-seq datasets using TRUST4, as described in "Language Model Embedding Classifiers Enable Identification of Multiple Sclerosis-Associated BCRs and Repertoires." Related publicationA detailed description of the database and its biological analyses is provided in the associated preprint: DOI: https://doi.org/10.64898/2026.03.05.709606 Data provenanceAll sequences originate from previously published and publicly available datasets. Original study references are included in the accompanying documentation.