RIPSeeker

RIPSeeker identifies RNA–protein interaction sites from RNA immunoprecipitation sequencing (RIP-seq) data to map genome-wide RNA transcripts that interact with specific proteins.

Key Features:

• De novo RIP peak prediction: Uses a two-state Hidden Markov Model (HMM) with negative binomial emission probabilities to infer and discriminate RIP peaks from RIP-seq alignments.
• Superior sensitivity and specificity: Empirical evaluations across three RIP-seq and two PAR-CLIP datasets involving six RNA-binding proteins reported an average area under the ROC curve of 0.80 and superior sensitivity and specificity relative to comparison methods.
• Biological validation: Identified peaks show significant enrichment for biologically meaningful genomic elements, published sequence motifs, and associations with canonical transcripts known to interact with the examined proteins.
• Post-alignment processing, visualization, and annotation: Integrated tools perform post-alignment processing, visualization, and annotation of RIP-seq data.

Scientific Applications:

• RIP-seq analysis: Identification of genome-wide RNA transcripts that interact with specific proteins from RIP-seq experiments.
• PAR-CLIP and related technologies: Applicable to related sequencing technologies such as PAR-CLIP for mapping RNA–protein interactions.
• Gene regulation and post-transcriptional modification studies: Support for analyses of gene regulation and post-transcriptional modifications mediated by RNA-binding proteins.
• Characterization of RNA-binding proteins: Facilitates discovery and characterization of functional roles of RNA-binding proteins and their transcript targets.

Methodology:

Computational methods explicitly include a two-state HMM with negative binomial emission probabilities applied to RIP-seq alignments, post-alignment processing, and validation by enrichment for genomic elements, published sequence motifs, and associations with canonical transcripts.