wavClusteR

wavClusteR is an R package for analyzing PAR-CLIP data, a next-generation sequencing technique that identifies transcriptome-wide interaction sites between RNA and RNA-binding proteins. PAR-CLIP introduces specific base transitions at RNA-protein interaction sites, serving as markers for identifying these sites. However, distinguishing these meaningful transitions from those caused by cell type-specific SNPs or sequencing errors is challenging and necessitates robust statistical methods to minimize false discoveries.

wavClusteR addresses this challenge by implementing a transition-centered algorithm powered by a Bayesian network approach. This approach assigns posterior log-odds to observed transitions, quantifying the confidence in RNA-protein interactions with high resolution.