# CoMBCR: Co-Learning Multi-Modalities of BCRs and gene expressions

## TL;DR

CoMBCR is a new tool that combines B-cell receptor and gene expression data to better understand B-cell biology and disease.

## Contribution

CoMBCR introduces a novel method for co-learning BCRs and gene expressions in a unified latent space.

## Key findings

- CoMBCR improves B-cell feature representation compared to methods using only BCRs.
- It reveals immune responses and CDR3 motif preferences in SARS-CoV-2-specific memory B cells.
- CoMBCR traces malignant B-cell development and uncovers survival patterns in lymphoma patients.

## Abstract

B-cell receptors (BCRs) and gene expression profiles are two distinct yet complementary modalities of B cells. However, most analyses treat them independently. Here, we present CoMBCR, a B-cell embedding tool that co-learns BCRs and gene expressions, representing data within a unified latent space for downstream analysis.

We applied CoMBCR to 126,791 B cells from diverse datasets with matched BCRs and gene expressions. First, CoMBCR outperforms the methods solely encoding BCRs in capturing B-cell biological features, achieving at least 0.1 improvement in Matthews Correlation Coefficient on a SARS-CoV-2 binding prediction task. Second, CoMBCR reveals active immune responses and CDR3 motif preferences through modality gap analysis in SARS-CoV-2-specific memory B cells. Moreover, when supported by spatial transcriptomics data, CoMBCR accurately traces the developmental trajectories of malignant B cells and uncovers transcriptional patterns associated with their survival within lymphoma patients.

The CoMBCR software is publicly available under the MIT License at https://github.com/deepomicslab/CoMBCR.git.

shuaicli@cityu.edu.hk