RAISE quantifies RNA binding protein activity via transcriptomic profiling

Alternative splicing (AS) greatly expands proteomic diversity and is tightly regulated by RNA-binding proteins (RBPs), whose activities reflect both the direction and the strength of their regulatory influence. However, inferring RBP activity remains challenging owing to the incomplete characterization of RBP–splicing regulatory relationships and the lack of experimental approaches for direct activity measurement. Here, we present RAISE, a computational framework that integrates multimodal evidence including CLIP-seq, sequence motifs, and splicing alterations derived from RNA-seq data to identify RBP targets, construct RBP–exon regulatory networks and quantitatively infer RBP activity across conditions. The targets identified by RAISE recapitulate the known functions and characteristic regulatory patterns of RBPs. RAISE accurately inferred the activity of diverse RBPs across multiple cell lines and perturbation experiments, and effectively captured activity changes independent of RBP expression levels. When applied to The Cancer Genome Atlas (TCGA) datasets, RAISE revealed RBPs with altered activity in cancer, underscoring its potential for large-scale analyses of splicing regulation. Furthermore, we provide RAISEDB, a database of predicted RBP–exon interactions. Collectively, RAISE enables systematic quantification of RBP activity and illuminates splicing regulatory networks across biological and disease contexts. Alternative splicing is regulated by RNA-binding proteins, but measuring their regulatory activity remains challenging. Here, the authors develop RAISE, a computational framework that integrates multimodal data to construct splicing regulatory networks and quantify activity from transcriptomic data.