B-cell receptor reconstruction from single-cell RNA-seq with VDJPuzzle

Abstract

Supplementary data are available at Bioinformatics online.

FB5P-seq: FACS-Based 5-Prime End Single-Cell RNA-seq for Integrative Analysis of Transcriptome and Antigen Receptor Repertoire in B and T Cells

“…As demonstrated by our studies of human tonsil B cells and peripheral blood antigen-specific T cells, FB5P-seq is particularly adapted for the integrative analysis of surface phenotype (through index sorting), gene expression, and antigen receptor repertoire of rare phenotypically defined B or T cells. Because FB5P-seq integrates three layers of barcodes (UMIs, cellular barcodes, and plate barcodes), the workflow is more cost-and time-effective than Smart-Seq2 (31), which has been so far the method of choice for plate-based scRNA-seq analysis of gene expression and antigen receptor repertoire in B or T cells (18,19,21,22). The molecular detection limit of FB5P-seq (10-20 molecules) was higher than what has been reported for Smart-seq2 (11) (seven molecules), suggesting that FB5P-seq may recover the expression of less genes per cell.…”

“…Experimental and computational approaches to infer TCR and BCR sequence from scRNA-seq datasets of T and B cells have been developed, relying either on data produced by plate-based full-length sequencing (Smart-seq2) (18)(19)(20)(21)(22) or droplet-based 5 ′ -end sequencing (10× Genomics) (23). The former allows for a deep analysis of phenotypically defined FACS-sorted cells but is costly, labor intensive, and does not support unique molecular identifiers (UMIs, random DNA barcodes incorporated into cDNA molecules during RT and enabling the quantitative deconvolution of scRNAseq reads despite potential cDNA amplification artifacts).…”

T‐cell receptor and B‐cell receptor repertoire profiling in adaptive immunity

“…Another powerful group of methods for studying repertoires are the single cell‐based approaches . These include methods that specifically target both chains of immune receptor molecules , as well as standard single‐cell transcriptomics analyses from which the sequences of immune receptors are extracted . In the first case, additional information on cell phenotype can be obtained either through separation by FACS sorting with specific surface markers or by targeted amplification of functional genes characteristic of different cell subsets .…”

Reconstructing B cell receptor sequences from short-read single cell RNA-sequencir with BRAPeS

“…To evaluate BRAPes, we first trimmed the original reads (50bp for the human data and 150bp for the mouse data) and kept only the outer 25 or 30 bases. We compared BRAPeS' performance on the trimmed data to two other previously published methods -BASIC (Canzar et al, 2017) and VDJPuzzle (Rizzetto et al, 2018) applied either on the trimmed data or the original long reads.…”

“…The random mutations make BCR reconstruction a challenging task. While methods to reconstruct BCR sequences from full length scRNA-seq are available (Canzar et al, 2017;Rizzetto et al, 2018;Lindeman et al, 2018) (as well as single cell V(D)J enriched libraries from 10x Genomics (Single Cell Immune Profiling -10x Genomics)), they were only tested on long reads (150bp and 50bp).…”