RNAct

RNAct provides genome-wide protein–RNA interaction predictions and integrates experimental datasets to map the protein–RNA interactome in human, mouse, and yeast.

Key Features:

• Extensive interaction database: Contains approximately 5.87 billion pairwise protein–RNA interactions computed with about 120 years of computation time on the Centre for Genomic Regulation (CRG) high-performance computing cluster.
• Integration of experimental and predictive data: Combines experimentally validated interactions (including ENCODE) with ab initio predictions to expand interaction coverage beyond available experimental datasets.
• Prediction method (catRAPID): Uses the catRAPID algorithm for ab initio protein–RNA interaction predictions.
• RNA-binding proteome coverage: Provides full coverage of the RNA-binding proteome, addressing the gap between ~250 proteins with experimentally validated RNA targets and ~1400 proteins with evidence of RNA-binding activity.
• Species coverage: Includes interaction data for human, mouse, and yeast genomes.

Scientific Applications:

• Mechanistic studies of RNA processing: Supports investigation of defects in RNA splicing, localization, and translation linked to protein–RNA interactions.
• Aggregation and disease research: Enables exploration of protein–RNA interactions implicated in aggregate formation and related pathologies.
• Discovery of novel interactions: Facilitates identification of previously uncharacterized protein–RNA interactions beyond experimental datasets.
• Comparative and evolutionary analyses: Allows cross-species comparisons (human, mouse, yeast) to study conserved interaction mechanisms.

Methodology:

Predictions were generated using the catRAPID ab initio method and integrated with experimental interaction data (including ENCODE); the database comprises ~5.87 billion pairwise interactions computed with ~120 years of CPU time on the CRG high-performance computing cluster.