Understanding how a single cell matures and takes on a specific role in the body is a major goal in biology and medicine. Thanks to the rise of single-cell RNA sequencing (scRNA-seq), scientists can now examine gene activity in individual cells to see how they change during development. But analyzing this data is tricky, cells are noisy and unpredictable, and traditional tools often struggle to clearly map how they transform into specialized types.

To tackle this, researchers at the Wuhan Institute of Technology developed a new method called scRDEN. This tool helps researchers identify distinct cell groups and uncover how groups of genes interact and shift during the process of cell differentiation.

The overall analysis of scRDEN encompasses three parts

(a) Motivation. Compared with the differential expression of individual genes in single-cell data, the differential expression of gene pairs is more stable and has stronger robustness during the cell differentiation process. (b) scRDEN Workflow. It demonstrates the entire process starting from the single-cell transcription matrix and ultimately achieving clustering analysis and trajectory inference. (c) Downstream Analysis. It includes the enrichment analysis of differentially expressed genes and the dynamic gene rank differential expression network identification along with the different directions of differentiation.

What makes scRDEN stand out is how it stabilizes the often messy data from individual cells. Instead of looking just at individual gene levels (which can be unreliable), scRDEN focuses on interactions between genes, making it easier to spot consistent patterns. It also combines results from different analysis techniques to improve accuracy.

When applied to several human and mouse datasets, scRDEN didn’t just identify cell types, it also revealed how key genes, like transcription factors and marker genes, increase or decrease in influence over time. These patterns offer new clues into how cells gradually develop stable identities. Especially impressive was scRDEN’s performance with a large and complex mouse brain dataset, where it successfully mapped out branching cell paths and regulatory changes.