Comprehensive characterization of V(D)J recombination from long-read transcriptomic data with VDJcraft

Area of Science:

• Immunology
• Genomics
• Bioinformatics

Background:

• V(D)J recombination is crucial for generating antigen receptor diversity in developing B and T cells.
• Long-read transcriptome sequencing offers higher accuracy for V(D)J event resolution compared to short-read methods.
• Existing analysis tools are not optimized for long-read data.

Purpose of the Study:

• To develop the first integrated pipeline, VDJcraft, for V(D)J recombination analysis using long-read transcriptome sequencing data.
• To improve the accuracy and sensitivity of V(D)J gene detection and recombination analysis.

Main Methods:

• VDJcraft employs a two-pass alignment strategy: global alignment with minimap2 and local realignment with IMGT.
• A customized module enhances D-gene detection sensitivity and positional precision.
• Consensus-based correction reduces sequencing errors, and IMGT motifs annotate antigen-binding regions.

Main Results:

• VDJcraft accurately recovered full-length V(D)J-C sequences, outperforming existing methods in gene detection and accuracy.
• Long-read VDJ calls showed significantly higher concordance with high-confidence short-read calls.
• Identified 31 novel gene subclasses and revealed distinct V(D)J recombination patterns in a COVID-19 patient.

Conclusions:

• VDJcraft provides a robust framework for V(D)J characterization using long-read sequencing.
• The pipeline enables the discovery of disease-associated immune signatures and novel immune gene subclasses.