May 31, 2026
Single-Cell GEX and VDJ (TCR) Sequencing Protocol: Library Preparation, Sequencing, and Analysis
- Hyojin Song1,
- Kiran Narta1,
- Joanna Pyczek1,
- Jennifer Chan1,
- Sorana Morrissy1
- 1University of Calgary
Protocol Citation: Hyojin Song, Kiran Narta, Joanna Pyczek, Jennifer Chan, Sorana Morrissy 2026. Single-Cell GEX and VDJ (TCR) Sequencing Protocol: Library Preparation, Sequencing, and Analysis. protocols.io https://dx.doi.org/10.17504/protocols.io.yxmvmdoebv3p/v1
Manuscript citation:
Zemp et al. 2026. GPNMB-directed CAR T-cell therapy against MiT/TFE family fusion-driven solid tumors. Nature Cancer. Accepted.
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: May 28, 2026
Last Modified: May 31, 2026
Protocol Integer ID: 318136
Keywords: profiling of peripheral blood mononuclear cell, peripheral blood mononuclear cell, cell transcriptomic, 10x genomics chromium gem, cell gex, downstream scrna, gcar1 status evaluation, marker gene profiling, human genome, cell receptor, sequencing protocol
Abstract
Comprehensive single-cell transcriptomic (GEX) and T-cell receptor (TCR/VDJ) profiling of peripheral blood mononuclear cells (PBMCs) using the 10x Genomics Chromium GEM-X platform. The workflow includes library preparation, sequencing, TCR clonotype assignment, and downstream scRNA-seq analysis with custom reference (human genome + GCAR1 construct), quality control, integration, clustering, cell-type annotation via SCimilarity, GCAR1 status evaluation, and marker gene profiling.
Guidelines
Notes
• All 10x Genomics library preparation steps must follow the manufacturer’s Chromium GEM-X Single Cell V(D)J + 5’ Gene Expression (v3) kit protocols exactly.
• The customized reference must include the full GCAR1 construct sequence for accurate detection of CAR-expressing cells.
• QC thresholds (especially mitochondrial content) may require optimization per sample type or donor.
• Integration and clustering parameters (resolution 0.8, rpca method) were chosen to yield 30 clusters; these may be adjusted based on dataset complexity.
• All analyses assume access to a high-performance computing environment suitable for cellranger and Seurat workflows.
• Include appropriate technical replicates and controls (e.g., healthy donor PBMCs) where possible.
Materials
Cryopreserved PBMC samples
Warm DMEM + 10% FBS (for thawing and washing)
1× PBS + 1% BSA (cell resuspension buffer)
10x Genomics Chromium GEM-X Single Cell V(D)J + 5’ Gene Expression (v3) kit
Procedure
Library Preparation. Thaw PBMC samples and wash with warm DMEM + 10% FBS. Perform automated cell counting. Resuspend cells in 1× PBS + 1% BSA to a final concentration of 100–2,000 cells/µL. Prepare Gene Expression (GEX) and T-cell receptor (TCR) libraries according to the manufacturer’s protocols for the 10x Genomics Chromium GEM-X Single Cell V(D)J + 5’ Gene Expression (v3) kit.
Single-Cell TCR (scTCR) Sequencing Analysis. Analyze VDJ data using scRepertoire (v2.0.7).1 Assign each cell a TCR clonotype based on paired TRA and TRB chains. Uniquely identify each TCR clonotype using a combination of V, D, J, C gene segments and CDR nucleotide sequences. Retain only cells with a single productive TRB chain for downstream analysis.
Single-Cell RNA (scRNA) Sequencing and Analysis
Pre-processing and Quality Control (QC). Generate a customized reference genome using cellranger mkref (v7.0.1) by aggregating the human reference (GRCh38/hg38) with GCAR1 construct sequences.2 Align sequencing data to the customized reference using cellranger multi (v7.0.1).2 Identify VDJ+ cells from the GEX dataset and subset them for downstream analyses. Perform all subsequent processing and QC in Seurat (v5).3 Identify GCAR1-expressing cells using subset (SeuratObject, subset = GCAR > 0). Apply the following QC filters:
- Raw nCount_RNA > 5,000
- Mitochondrial RNA percentage (percent.mt) < 10% (except for the GCAR1 Product sample, wherepercent.mt > 20% is allowed)
Normalize data using SCTransform, regressing out mitochondrial content. Integrate all samples across timepoints: select integration features with SelectIntegrationFeatures(), identify anchors using FindIntegrationAnchors(reduction = "rpca"), and generate integrated expression values with IntegrateData(). Perform cell clustering at resolution 0.8 (resulting in 30 clusters).
Cluster-Level Cell Type Predictions. Perform unsupervised cell-type labelling using SCimilarity (v0.2.0; “unconstrained” model).4 Re-run the model in “constrained” mode to obtain a min_dist value for each cell’s label prediction (0 indicates an identical match to the reference). Calculate a cluster label score that incorporates the min_dist value per label (x),and the number of cells with each label (xc), per cluster, as:
cluster label score_x = (threshold(min_dist) – median(min_dist))_k × proportion(x_c)
Validate annotations. Perform pathway enrichment analysis to extract cluster-specific marker genes with FindAllMarkers() (using performNormalization()), filter for markers with avg_log2FC > 1.0, and query with the g:Profiler2 R package focusing on GO:BP terms.5,6
Cluster-Level GCAR1 Status. Annotate clusters as GCAR1+ if they contain >250 cells or >15% GCAR1+ cells. Calculate p-values for the proportion of GCAR1-positive cells in each cluster using a two-proportion z-test.
Cell-Level Cell Type Classification. Refine SCimilarity cell-level labels into broad cell types using semi-manual curation and tumour marker gene expression levels. Retain cells only if cell-level and cluster-level annotations agree on T-cell status (to eliminate ambiguous assignments).
Marker Gene Expression Profiling. Profile CAR T-cell immunophenotyping genes using established published marker lists.7
Protocol references
1. Borcherding, N., Bormann, N. L. 26 Kraus, G. scRepertoire: An R-based toolkit for single-cell immune receptor analysis. F1000Res. 9, 47 (2020).
2. Zheng, G. X. Y. et al. Massively parallel digital transcriptional profiling of single cells. Nat. Commun. 8, 14049 (2017).
3. Hao, Y. et al. Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat. Biotechnol. 42, 293–304 (2024).
4. Heimberg, G. et al. Scalable querying of human cell atlases via a foundational model reveals commonalities across fibrosis-associated macrophages. Preprint at https://doi.org/10.1101/2023.07.18.549537 (2023).
5. Kolberg, L., Raudvere, U., Kuzmin, I., Vilo, J. 26 Peterson, H. gprofiler2 — an R package for gene list functional enrichment analysis and namespace conversion toolset g:Profiler. F1000Res. 9, 709 (2020).
6. Raudvere, U. et al. g:Profiler: a web server for functional enrichment analysis and conversions of gene lists (2019 update). Nucleic Acids Res. 47, W191–W198 (2019).
7. Anderson, N. D. et al. Transcriptional signatures associated with persisting CD19 CAR-T cells in children with leukemia. Nat. Med. 29, 1700–1709 (2023).