sciCSR infers B cell state transition and predicts class-switch recombination dynamics using single-cell...

Area of Science:

• Immunology
• Computational Biology
• Genomics

Background:

• Class-switch recombination (CSR) is a critical process in B cell maturation, essential for adaptive immunity.
• Analyzing CSR dynamics at a single-cell level provides deeper insights into B cell development and immune responses.
• Current methods for analyzing CSR from single-cell RNA sequencing (scRNA-seq) data have limitations.

Purpose of the Study:

• To develop and validate sciCSR, a novel computational pipeline for inferring CSR events and dynamics from scRNA-seq data.
• To differentiate productive immunoglobulin heavy-chain transcripts from germline transcripts for accurate CSR analysis.
• To model B cell CSR dynamics using a Markov state model.

Main Methods:

• sciCSR pipeline re-analyzes scRNA-seq alignments to distinguish productive and germline immunoglobulin transcripts.
• A Markov state model is constructed from scRNA-seq data to infer CSR dynamics and direction.
• The pipeline was validated using simulated data and real scRNA-seq data from SARS-CoV-2 vaccination time-course experiments.

Main Results:

• sciCSR accurately differentiates productive and germline immunoglobulin transcripts.
• The computational pipeline successfully predicts B cell receptor isotype distribution with high accuracy (cosine similarity ~0.9) from earlier time points.
• sciCSR identifies B cell-specific transitions missed by conventional RNA velocity analyses.

Conclusions:

• sciCSR is a validated computational tool for analyzing CSR events and dynamics in B cells using scRNA-seq.
• The pipeline provides accurate predictions of isotype distribution, aiding in the understanding of immune responses.
• sciCSR enhances the analysis of B cell maturation and CSR, offering novel insights into immune dynamics.