Benchmarking REFLEX for Scalable Single-Cell TCR and Transcriptome Profiling

Abstract

Single-cell RNA sequencing (scRNA-seq) combined with immune receptor analysis is a significant breakthrough in immunology, enabling detect of gene expression and T-cell receptor (TCR) at the single-cell level in the same sample. However, current "gold standard" methods like 10x Genomics 5' V(D)J rely entirely on fresh samples, which severely limits their application in clinical studies requiring multi-period sample collection and preservation, and are also extremely costly. To address this challenge, the "REFLEX" technology was recently developed. This technology base on 10x FLEX platform with fixed samples, enables sequencing the TCR and transcriptome in fixed cells together, by introducing custom probes targeting the constant and variable regions of the TCR mRNA chains. The primary purpose of this study is to independently validate and benchmark the performance of the REFLEX workflow in controlled cell line models. Furthermore by mixing Jurkat T cells into peripheral blood mononuclear cells (PBMCs) at different gradients (Spike-in), we are able to compare the accuracy, sensitivity, and reproducibility of REFLEX with the conventional 5' V(D)J method. The results showed that REFLEX exhibited extremely high sequence purity. In three biological replicates of Jurkat cell models, the proportion of TRA and TRB sequences matching the consensus without any mismatch (0\,bp) reached approximately 96\%, with a double-strand matching rate of approximately 68\%. Compared to fresh 5' samples, REFLEX can capture a greater number and type of cells, and it also outperforms conventional FLEX in terms of UMI and gene detection sensitivity. Although the proportion of cells from which a complete TCR sequence was recovered was lower in REFLEX than in the 5' V(D)J method compared to the 5' method, it successfully achieved simultaneous mapping of endogenous TCR diversity and gene expression in complex backgrounds.

In summary, this study validates that REFLEX provides an economical and highly scalable alternative to TCR sequencing, establishing necessary quality control benchmarks for future analysis of complex, fixed samples such as solid tumor microenvironments.

Popular Abstract

Why do some people have immune systems that are superior to others---able to fight infections, suppress cancer, or respond better to treatment? This largely due to their genes, specifically the genes of their T cells. T cells are a type of white blood cell that carries a unique receptor on their surface called the T-cell receptor (TCR). Each T cell has a slightly different version of the TCR, and this difference determines which substances it can recognize. By sequencing the TCR sequences, researchers can understand how the immune system fights disease.

However, classic TCR sequencing methods require fresh, live cells. Samples need to be processed as quickly as possible, which is feasible in the laboratory but a challenge in clinical settings, therefore most existing sample use formaldehyde to preserve samples. While this preservation breaks the RNA strands in cells, altering their structure and making them unanalyzable using classic sequencing methods.

REFLEX was developed to address this problem. Instead of requiring a complete RNA strand, it uses small probes to bind to specific RNA sequences. This method allows for the simultaneous processing of up to 16 samples, significantly reducing costs and eliminating batch variability, enabling study of samples collected across multiple time or locations.

The purpose of this project is to independently test the REFLEX method. To achieve this, I used Jurkat cells, a laboratory-immortalized T cell type with a known TCR sequence, and mixed them into normal human blood samples at concentrations ranging from 10\% to 0.5\%. The aim was to answer the following questions: How accurately can REFLEX capture TCR sequences, and what are its detection limits?

We designed three independent experiments and obtained the following results: REFLEX captured TCR sequences with approximately 96\% accuracy, and the method was still able to detect signals even at the lowest tested concentration. Comparing REFLEX with traditional sequencing methods, we found that REFLEX captured a larger number of cells and had higher gene detection sensitivity. However, it uses fixed cells, the proportion of total TCR sequences captured per cell is lower, but the proportion of both $\alpha$ and $\beta$ TCR sequences per cell is higher.

This means that researchers can use blood samples collected from cancer patients months or years ago for immunological analysis and have the ability to track changes in the immune system during treatment, without being limited by the need for fresh samples required by traditional methods.

• author: Fan, Shiming
• supervisor: Julia Bräunig; Paul Theodor Pyl
• organization: Department of Immunotechnology; Educational programmes, LTH
• course: KIMM01 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: scRNA-seq, TCR, REFLEX, FLEX, fixed RNA profiling, Jurkat, PBMC, CDR3, V(D)J sequencing, multiplexing, R, Seurat
• language: English
• id: 9241941
• date added to LUP: 2026-07-14 21:50:51
• date last changed: 2026-07-14 21:50:51