New Pipeline Sharpens Single-Cell Profiling of CAR Immune Therapies

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Apr 29 2026

A new CAR-aware pipeline could help researchers track engineered immune cells more precisely, improving how complex cell therapies are studied, compared, and refined.

Study: CERTOMICS: trusted single-cell multiomics pipeline for high-resolution profiling of adoptive cellular immunotherapies. Image Credit: Alpha Tauri 3D Graphics / Shutterstock

In a recent study published in the journal Bioinformatics, a group of researchers developed a scalable and reproducible computational pipeline for accurate identification and comprehensive profiling of chimeric antigen receptor (CAR)-engineered immune cells using integrated single-cell multiomics data.

CAR-T Cell Profiling and Computational Gaps

Cellular immunotherapies, including CAR T cell therapies, have transformed treatment for several hematological cancers by using engineered immune cells to specifically target and eliminate cancer cells.

Despite their promise, CAR T cell therapies still face challenges, including high costs, disease relapse, antigen escape, severe side effects, and suppression by the tumor microenvironment, particularly in solid tumors.

A better understanding of cellular heterogeneity and the molecular profiles and cellular interactions that shape therapeutic response is critical to identify strategies to overcome these challenges. 

Current computational workflows lack standardization and CAR-specific functionality, and the lack of reliable methods for identifying CAR-positive cells underscores the need for dedicated analytical tools to facilitate more accurate interpretation of these data and optimize CAR T cell therapies.

CERTOMICS Rationale and Pipeline Design

Recently developed immunotherapies, including CAR T cell therapies, act as “living drugs,” in which engineered immune cells continually engage tumor cells and the host immune system.

Understanding how CAR-engineered cells respond, persist, expand, and interact with host immune and tumor cells needs single-cell analysis, but existing computational tools do not meet the specific needs of CAR-engineered cells.

To address these limitations, CERTOMICS was developed as a Nextflow-based computational pipeline designed specifically for single-cell multiomics profiling of CAR-engineered immune cells. It integrates CAR-specific detection, multi-modal data processing, and standardized workflows into a unified analytical framework.

CERTOMICS supports the integration of multiple data modalities, including gene expression (GEX), immune receptor sequencing V(D)J, and antibody-derived tags (ADT). This multi-modal capability allows a more comprehensive analysis of immune cell populations and their functional states.

A main feature of CERTOMICS is its CAR-aware reference construction. Because CAR transgenes are synthetic and not included in standard reference genomes, the pipeline incorporates user-provided CAR sequence and annotation files to build customized references.

The pipeline is organized into three main components. The first part is a reference-handling component, HANDLE_REFERENCES, that produces CAR-aware reference genomes for the sequenced sample.

The second component is called the core analysis module, RUN_SECONDARY_ANALYSIS, that analyzes the multi-modal data, generates integrated results, and provides annotated output.

The third component is an optional quality control module, RUN_QUALITY_CONTROL, that evaluates sequencing quality using established tools.

Overview of the CERTOMICS pipeline for single-cell multiomics analysis including CAR T cell products. A set of samples (Sample1 to SampleN), each with associated sequencing libraries - gene expression (GEX), T cell receptor profiling [V(D)J], and antibody-derived tags (ADT) - as well as CAR construct data (FASTA and GTF) are processed. CERTOMICS first builds a CAR-aware reference (HANDLE_GEX_REFERENCE), processes multi-modal sequencing data (RUN_SECONDARY_ANALYSIS) into a merged, annotated Seurat object, and generates an interactive results webpage including CAR-metrics. Users can optionally enable RUN_QUALITY_CONTROL to perform multi-modal QC via FastQC, FASTQ_Screen, and MultiQC. The pipeline supports three execution modes (--reference, --analysis, and --full), allowing modular use of individual components. Contribution to the CAR resource (GitHub symbol), extensibility of the pipeline, and interoperability with external tools are highlighted.

CAR-Specific Detection and Multiomics Reporting

A distinguishing feature of CERTOMICS is its ability to detect and quantify CAR-positive cells using specialized quality control metrics. These metrics have two levels: read-level metrics assess sequencing coverage across CAR constructs and evaluate transgene expression.

Count-level metrics quantify CAR-positive cell frequencies across immune cell populations and conditions, enabling comparisons before and after CAR expansion and with CAR-negative controls.

There is an optional validation step within the pipeline that allows for confirmation of the CAR construct's identity through sequencing read comparisons to alternative constructs. This improves the specificity of the data, ensuring that the engineered CAR cells are accurately identified.

Additionally, CERTOMICS produces full reports that include gene expression statistics, immune receptor diversity, and clonotype composition. The outputs provide an understanding of how much the immune cells have expanded, their functional state, and changes in their immune repertoire, which may inform downstream analysis of their interactions with tumor and host immune cells.

CERTOMICS can create an interactive webpage that lets users view multiple data sources, sample types, and results on a single page. The report provides a visual representation of several parameters related to CAR metrics, sequencing quality, and the characteristics of the immune repertoire.

In addition, combining multi-modal data into a unified output allows researchers to look at the relationships within the three categories of gene expression, protein markers, and immune receptor sequences. 

A key output is a merged Seurat object that consolidates multi-modal data across samples and adds quality-control metadata, cell-type annotations, and clonotype information for downstream analysis.

To support CAR detection and analysis, CERTOMICS includes a curated repository of CAR construct sequences and annotations. This resource contains nucleotide sequences and domain information for multiple CAR T cell products, enabling accurate reference construction and analysis.

This repository helps researchers manage diverse CAR designs and limited public sequences. Users can also contribute new constructs, ensuring that the resource remains up to date with emerging therapies.

Together, this standardized pipeline and database improve the reproducibility and scalability of single-cell studies.

Implementation and Future Applications of CERTOMICS

CERTOMICS is implemented using Nextflow, a workflow management system that supports reproducibility, scalability, and portability across computational environments. The pipeline can be executed locally or on high-performance computing systems using workload managers such as Slurm. This modularity also helps run different types of research projects, such as creating references or performing multi-modal analyses.

The authors note that CERTOMICS is currently optimized for widely used 10x Genomics protocols, while future versions may support emerging assays and engineered immune-cell platforms.

CERTOMICS addresses a critical gap in computational tools for CAR-engineered cell analysis by combining multi-modal data integration with CAR-specific functionality. Current pipelines are limited because they lack support for multiomics data or fail to incorporate CAR-specific features.

By enabling precise detection of CAR-positive cells and integrating diverse data types into a single framework, CERTOMICS provides a comprehensive platform for studying cellular heterogeneity and therapeutic mechanisms. Its standardized design facilitates reproducible analyses and may support future patient-specific approaches, including virtual twin analyses, although clinical utility would require further validation.

Conclusion

CERTOMICS represents a significant advancement in the computational analysis of adoptive cellular immunotherapies by providing a standardized, scalable, and CAR-aware pipeline for single-cell multiomics data.

By integrating gene expression, immune receptor profiling, and protein data, the pipeline is designed to improve the accuracy, reproducibility, and interpretability of characterizing engineered immune cells and to support downstream studies of their interactions with tumor and host immune cells. There are also CAR-specific references constructed for use within CERTOMICS, which provide supporting quality control metrics.

The modular and extensible structure of CERTOMICS enables future technological developments and therapeutic approaches to be integrated into CERTOMICS. Overall, CERTOMICS provides a better understanding and allows for more standardized investigation of CAR-based therapies and next-generation immunotherapies.

Download your PDF copy by clicking here.