Single-cell RNA sequencing allows researchers to measure gene activity in individual cells, making it possible to study how different cell types respond to disease, treatment or genetic changes. But generating the sequencing data is only the beginning. Researchers often need several separate analyses to understand what the results mean biologically.

These can include identifying genes that change between groups, determining which biological pathways are affected, examining gene regulatory networks and studying how different cell types communicate with one another.

Researchers at the Department of Biomedical Science and Engineering at Konkuk University developed scFlowReport, a workflow designed to bring these downstream analyses together under a single comparison framework.

scFlowReport accepts a processed, cell-type-annotated Seurat object (.rds) together with a YAML configuration file as input and executes a three-stage workflow comprising preprocessing, analytical module execution, and report generation. The analytical stage includes cell-level and pseudobulk differential expression analysis, transcription factor regulon analysis, and cell–cell communication analysis. Functional enrichment analysis is subsequently performed. scFlowReport produces a standardized set of outputs, including a portable static HTML report, standardized result tables, and publication-quality figures.

The challenge of analyzing single-cell RNA sequencing data

Many single-cell RNA sequencing projects are built around comparisons, such as disease versus healthy controls or treated versus untreated samples.

After cells have been identified and classified, researchers may use several different software tools to investigate the biological differences between the groups. Each tool can require its own settings, input formats and scripts.

This makes it difficult to ensure that exactly the same comparison is being used throughout the analysis. Results from the different tools also need to be organized and interpreted together.

scFlowReport was developed to simplify this process.

One comparison across multiple analyses

scFlowReport begins with an already processed and cell-type-annotated Seurat object, a commonly used format for single-cell RNA sequencing analysis.

The user defines one comparison, such as disease versus control, and scFlowReport carries that definition consistently through several downstream analyses.

These include differential gene expression at both the individual-cell and pseudobulk levels, functional enrichment analysis, transcription-factor regulatory network analysis using SCENIC and cell-cell communication analysis using LIANA.

Pseudobulk analysis combines gene expression measurements from cells belonging to the same biological sample before comparing groups. This can help account for the fact that cells collected from the same person or sample are not truly independent observations.

Connecting different layers of biology

Each analysis performed by scFlowReport provides a different view of the RNA sequencing data.

Differential expression identifies genes whose activity changes between groups. Functional enrichment connects those genes with biological processes and pathways. Regulatory analysis examines transcription factors that may help control gene activity, while cell-cell communication analysis investigates signaling between different cell populations.

Instead of requiring researchers to run these analyses separately and combine the outputs manually, scFlowReport coordinates them within one workflow.

The software then produces standardized figures, summary tables and a self-contained HTML report that can be viewed and shared without running a dedicated server.

Testing the workflow with atopic dermatitis data

The researchers demonstrated scFlowReport using a publicly available single-cell RNA sequencing dataset comparing atopic dermatitis with healthy skin.

The analysis included keratinocytes, fibroblasts, T cells and dendritic cells. Using the same disease-versus-control comparison across its different modules, scFlowReport generated differential expression, functional enrichment, regulatory network and cell communication results.

The workflow identified biological signals consistent with known inflammatory and epithelial changes in atopic dermatitis, demonstrating how several complementary analyses can be brought together in a consistent framework.

Making downstream analysis more reproducible

scFlowReport does not replace the individual analytical methods it incorporates. Instead, its main role is to coordinate established tools and ensure that they use the same comparison definitions.

This can reduce the amount of custom scripting required and make it easier to reproduce an analysis later or share results with collaborators.

As single-cell RNA sequencing datasets become larger and more complex, workflows that organize multiple types of downstream analysis may help researchers move more efficiently from gene expression measurements to biological interpretation.

Availability – The scFlowReport source code, execution instructions, and complete software environment information are available at https://github.com/ibclab-hub/scFlowReport