CRIT

CRIT identifies RNA-binding proteins (RBPs) that regulate circular RNAs (circRNAs) by applying non-negative matrix factorization to integrated multi-omics and genomics data to reveal regulators of circRNA biogenesis and modulation, with emphasis on cancer contexts.

Key Features:

• Non-negative matrix factorization: Uses non-negative matrix factorization as a core algorithm to decompose integrated datasets for regulator discovery.
• Multi-omics integration: Integrates various omics and genomics data alongside functional annotations to improve RBP identification.
• CLIP database validation: Analyzes circRNA–RBP interactions using two extensive cross-linking immunoprecipitation (CLIP) databases to validate predictions against experimental binding data.
• RNA-seq knockdown analysis: Analyzes RNA sequencing (RNA-seq) datasets from shRNA/siRNA knockdown experiments to identify RBPs that influence global circRNA expression.
• Discovery of novel RBPs: Reported identification of 73 novel RBP regulators across thousands of samples.
• Splicing and cancer associations: Highlights that authentic RBP regulators are often core splicing proteins and peripheral factors that exhibit alterations across diverse cancers.

Scientific Applications:

• RBP regulator discovery: Identification and prioritization of RBPs involved in circRNA biogenesis and modulation.
• Validation of candidate regulators: Cross-validation of predicted RBPs with CLIP data and RNA-seq knockdown responses.
• Cancer biology: Characterization of RBP alterations and circRNA regulatory networks across cancer datasets.
• Target prioritization: Provision of candidate RBPs for downstream experimental follow-up and potential therapeutic investigation.

Methodology:

Computational methods explicitly stated include non-negative matrix factorization on integrated multi-omics and genomics data with functional annotations, analysis of circRNA–RBP interactions using two CLIP databases, and analysis of RNA-seq datasets from shRNA/siRNA knockdown experiments.