Crinet

Crinet infers genome-wide competing endogenous RNA (ceRNA) interactions and ceRNA groups by integrating expression datasets and applying network deconvolution to account for expression abundance sufficiency, collective regulation, and groupwise ceRNA effects.

Key Features:

• Implementation: Implemented in R for computational inference of genome-wide ceRNA networks.
• Inclusive ceRNA consideration: Incorporates messenger RNAs (mRNAs), long non-coding RNAs (lncRNAs), and pseudogenes as candidate ceRNAs.
• Expression abundance and collective regulation: Considers expression abundance sufficiency, collective regulation, and groupwise effects when evaluating ceRNA interactions.
• Network deconvolution: Employs a network deconvolution method to remove spurious ceRNA pairs and refine inferred interactions.
• Validation mechanisms: Validated using breast cancer data from The Cancer Genome Atlas (TCGA) with protein expression-based benchmarks and gene expression changes from knockdown assays.
• Hub and group identification: Identifies hub genes and ceRNA groups, including known suppressor and oncogene lncRNAs in breast cancer.

Scientific Applications:

• Regulatory circuit analysis in cancer: Infers ceRNA networks to study miRNA-mediated competition and its influence on gene expression patterns in diseases such as breast cancer using TCGA data.
• Non-coding RNA functional insight: Highlights roles of lncRNAs and pseudogenes within ceRNA groups, including known suppressor/oncogene lncRNAs relevant to cancer biology.
• Target discovery and prioritization: Reveals ceRNA-mediated regulatory relationships that can aid identification of candidate therapeutic targets and biomarkers.

Methodology:

Implemented in R; integrates comprehensive expression datasets including mRNAs, lncRNAs, and pseudogenes; and applies network analysis with network deconvolution to infer and refine genome-wide ceRNA interactions and groups.