Bridging Sequence–Structure Motifs and Genetic Variants for Genome-wide Dynamic RNA–Protein Interaction Profiling

We assembled a collection of 261 RNA–RBP interaction datasets across six human cell lines (K562, HepG2, HEK293, HEK293T, HeLa, and H9), encompassing 172 RBPs profiled using a uniform flagmarked protocol. The dataset comprises 65 CLIP-seq experiments (61 RBPs) obtained from the POSTAR repository and 196 eCLIP experiments (111 RBPs) from the ENCODE project. These unified datasets provide the basis for model training and evaluation across diverse cellular contexts.