Sequence variations of immune receptors

This has been a research direction I pursue since my post-doc days. I conduct large-scale computational analysis of the in vivo antibody repertoire and data from in vitro studies of B cell biology.

This has resulted in a number of bioinformatics tools to analyse antibody sequence variations using data generated with high-throughput sequencing methods. We analyse variations in terms of CSR, somatic hypermutation (SHM) and other diversification mechanisms observed in variable regions of antibodies.

Code

ImmunoMatch: A fine-tuned antibody language model for the identification of cognate paired immunoglobulin Heavy (H) and Light (L) chains. Trained from single-cell paired H-L amino acid sequences, we showed that the ImmunoMatch model reveals insights into how chain pairing propensity changes across B cell maturation stages and explored how this can be game-changing in antibody discovery pipelines from spatial transcriptomics datasets. (Nature Methods 2026)

BrepConvert: R package for flexible and detailed annotation of gene conversion events in immunoglobulin repertoire. Designed to analyse repertoire data from primates such as chicken, rabbit etc where gene conversion is prevalent to generate antibody diversity. (Discovery Immunology 2023, Immuongenetics 2026)

BrepPhylo: R package for fast lineage reconstruction, enabling analysis on large samples of clonotypes from BCR repertoire data. (Frontier in Immunology 2022)