CASREL Maps Cell‑Specific RNA Splicing from Single‑Cell Data with Machine Learning
CASREL is a new machine‑learning framework that reconstructs the regulatory links between RNA‑binding proteins and alternative splicing directly from single‑cell RNA sequencing (scRNA‑seq) data. By avoiding reliance on pre‑existing protein‑RNA binding annotations, it can uncover cell‑type‑specific splicing programs that traditional bulk assays miss.
Key points
- CASREL infers RNA‑binding protein–splicing links from scRNA‑seq without prior annotations.
- It uses ensemble learning plus SHAP for interpretable, cell‑type‑specific predictions.
- Validated across tissues, it matches known biology and uncovers new regulatory circuits.
The method combines ensemble learning with Shapley additive explanations (SHAP) to identify which RNA‑binding proteins drive specific isoform choices. It exploits unique features of scRNA‑seq, such as polarized isoform usage and reduced averaging across cells, making it robust to expression noise and highly interpretable.
Across a range of tissues and cell types, CASREL demonstrates strong accuracy and biological relevance, revealing putative regulatory circuits that align with known biology while also proposing novel protein‑splicing relationships. The approach opens a path to physiologically relevant, high‑resolution mapping of splicing regulation that could inform disease mechanisms and therapeutic targets.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us.
Comments
via GitHub Discussions