Skip to content

CERTOMICS

What is CERTOMICS

This Nextflow-based pipeline offers enhanced certainty in immunophenotyping and data interpretation, tailored for single-cell multiomics profiling of adoptive cellular immunotherapies. CERTOMICS enables comprehensive immune profiling by supporting single-cell 10x Genomics technologies, such as gene expression (GEX) sequencing, cell surface protein detection, and V(D)J (TCR and BCR) sequencing. To reliably characterize CAR-positive cells, the pipeline enables CAR detection based on a custom reference and incorporates specialized quality control metrics to ensure robust identification of CAR-positive cells.

Abb_CAR_T_cell_profiling_V2.png

Pipeline Overview

The pipeline is structured into three main processes: handling 10x data and libraries, performing core secondary analysis, and running quality control. The key feature of the pipeline is its ability to detect CAR-positive cells by incorporating a dedicated reference processing step. For this step, the user simply provide a CAR sequence file .fasta and CAR annotation file .gtf, for which CAR+ cells should be detected. For detailed explanation see: References. Common CAR sequences and annotation information can be found in our Resource.

  • Workflow: BUILD_CUSTOM_REFERENCE - Generates custom reference files for CellRanger based on the sequencing libraries (GEX, VDJ, ADT) and the CAR construct (FASTA, GTF) given.
  • Workflow: SECONDARY_ANALYSIS - Executes CellRanger multi and generates a merged, annotated Seurat object, and computes CAR-specific quality control metrics, with results summarized in a dedicated webpage.
  • Workflow: QUALITY_CONTROL - Conducts quality control assessments using FASTQC (v0.12.1) FASTQ Screen (v0.15.3), and a custom multi-modal MultiQC (v1.24.1) module evaluate quality and composition of a multi-modal single-cell sequencing experiment.

CAR-Specific Features

CERTOMICS offers several features specifically tailored to CAR-engineered cell products:

1. 🧬 CellRanger Multi

CERTOMICS supports processing diverse combinations of gene expression, V(D)J, and feature-barcode libraries across multiple samples.
It is fully compatible with 10x Genomics immune-profiling workflows using CellRanger Multi. For more information see CellRanger Output.

2. 📊 Multi-Modal QC

A custom multi-modal MultiQC module enables cross-sample evaluation of sequencing quality and modality integration. Additionall quality control assessments can be done using FastQ-Screen and FastQC of sequencing data. For more information see Quality Control parameters.

3. 🔍 CAR Transgene Detection

CERTOMICS includes a dedicated reference-processing step that identifies CAR-positive cells based on transgene expression. For more information see Reference building.

4. 🧪 Merged Seurat Object

Provides a merged Seurat object containing multi-modal single-cell data (RNA, VDJ and ADT assays) along with extensive metadata on cell type identity, quality metrices, cell cycle, and clonotype information, which serves the purpose of enabling detailed characterization and quality assessment of diverse cell populations from multiple samples. For a detailed explanation please see: Seurat Output

5. 🌐 Interactive CAR-Specific QC Reports

CERTOMICS produces an interactive HTML report (see Example) summarizing:

  • Read-level metrics include coverage plots across the CAR construct, enabling assessment of read distribution and sequencing protocol performance (5' or 3'), as well as absolute read counts per sample to evaluate transgene expression levels. Additionally, an optional validation step can be performed to con rm correct CAR construct identity by comparison to alternative CAR constructs, serving as internal controls to assess detection speci city.
  • Count-level metrics provide biological insights by quantifying CAR-positive cell frequencies at different immune cell populations. This allows the assessment of patient-specific variability or comparison across different condition, e.g. before vs. after CAR expansion, as well as a negative control from CAR-negative cell populations.

In addition to CAR-specific quality control, the webpage presents general GEX-specific statistics, including cell proportions, as well as VDJ sequencing statistics, such as clonotype composition for both T and B cells. These additional metrics provide insights into TCR/BCR diversity, expansion patterns, and immune repertoire changes, which are critical for understanding persistence, functionality, and immune cell interactions.

CAR__metrics

Quickstart guide

Downloading the pipeline

To install the pipeline, simply download the Git repository via HTTP.

git clone https://github.com/fraunhofer-izi/CERTOMICS.git

Before attempting to run the pipeline, ensure that Nextflow is installed and up-to-date (version >= 24.10.2). If you use the pipeline with the Singularity containers (recommended via-profile sigularity), Singularity must be installed and included in PATH, as well.

Supplying data

To provide all the necessary data (references, samples, etc.) a separate file is used. All values are captured in the YAML (or JSON) format in a so-called parameters file (or params-file) and passed to the pipeline with the -params-file flag. The parameters file should look something like this:

gene_expression_reference: "/path/to/gex/reference"
vdj_reference: "/path/to/vdj/reference"
samples:
  - name: 'sample_1'
    libraries:
      - fastq_id: 'sample_1_R'
        fastqs: "/path/to/sample_1_R"
        feature_types: 'Gene Expression'
      - fastq_id: 'sample_1_B'
        fastqs: "/path/to/sample_1_B"
        feature_types: 'VDJ-B'

A full explanation of the options you have and how to populate your own params-file can be found here.

Documentation

To get a better understanding of what you can do with the pipeline and the options you have, check out the documentation