CircAI: a comprehensive database of CircRNA associated with A-to-I RNA editing

Yulan Wang, Lingxiao Zou, Jian Zhao, Jing Wu, Meng Zhang, Jingjing Liu, Quan Wang, Xuejiang Guo, Xiaofeng Song, Yixuan WangDatabase J. Biol. Databases Curation 2025 · DOI

TL;DR

CircAI is a novel, comprehensive database that integrates multispecies circRNA data with A-to-I RNA editing information and functional predictions. This resource aims to bridge the gap in understanding the functional impact of RNA editing on circRNAs, offering a valuable tool for researchers investigating their roles in various biological processes and diseases.

What Problem Does This Paper Address?

Circular RNAs (circRNAs) are a fascinating class of noncoding RNAs with diverse regulatory functions, implicated in human diseases. RNA editing, particularly adenosine-to-inosine (A-to-I) editing, is a prevalent post-transcriptional modification that can alter RNA sequences and consequently their functions. While numerous databases exist for circRNAs and RNA editing events separately, there's a significant lack of integrated resources that specifically link these two phenomena and explore their functional consequences. Existing databases often have limitations in species coverage, functional annotation, or the integration of circRNA-specific editing data. This gap hinders researchers from fully understanding how RNA editing impacts circRNA structure, interactions, and ultimately, their roles in cellular processes and disease pathogenesis.

Methods

The CircAI database was constructed through a multi-step process:

  • CircRNA Identification: CircRNAs were systematically identified using the CIRI2 pipeline, with full-length sequences reconstructed using CIRI Full and circAST pipelines. A reference-based method was employed for circRNAs not fully assembled by these tools.
  • RNA Editing Site Detection: REDITools was used to detect A-to-I RNA editing sites. Reads were aligned to reference genomes, and stringent filtering criteria were applied to ensure high confidence. Known SNPs were excluded to minimize false positives.
  • Functional Annotation:
    • Coding Potential: Internal ribosomal entry sites (IRES) and open reading frames (ORFs) were predicted using IRESfinder and ORFfinder to assess the impact of editing on potential protein-coding capacity.
  • miRNA-circRNA Interactions: TargetScan and miRanda were used to predict miRNA targets on both edited (ET) and unedited (UT) circRNA sequences, allowing for the identification of "gain" or "loss" of interactions.
  • RNA Secondary Structure: RNAfold was employed to predict local RNA structures and quantify changes in RNA stability induced by editing.
  • RNA Editing Quantitative Trait Loci (edQTL): MatrixEQTL was used to link RNA editing levels with genotypes, identifying proximal SNPs associated with editing frequencies.
  • Database Integration: All collected data was integrated into a comprehensive database with a user-friendly web interface.

Datasets

CircAI was built using 533 RNA-seq datasets spanning eight species: Homo sapiens, Mus musculus, Macaca mulatta, Gallus gallus, Rattus norvegicus, Oryctolagus cuniculus, Sus scrofa, and Danio rerio. The data was sourced from the NCBI SRA database, with search terms 'RNA-seq' and 'RNase R' used to acquire relevant samples. Standardized reference genome versions, GTF annotation files, and RepeatMasker repeat annotation files were sourced from the UCSC Genome Browser. SNP data was obtained from dbSNP V151.

Key Insights and Ideas

  • Bridging the Gap: The most significant contribution is the creation of a unified platform that directly links circRNAs and RNA editing, providing a much-needed resource for functional exploration.
  • Multispecies Coverage: The inclusion of eight species significantly expands the scope beyond human-centric databases, allowing for cross-species comparisons and a broader understanding of evolutionary conservation and divergence in circRNA editing.
  • Comprehensive Functional Predictions: By integrating predictions for coding potential, miRNA interactions, secondary structures, and edQTLs, CircAI moves beyond simply cataloging events to offering insights into their functional consequences.
  • Focus on Noncoding Regions: The database specifically addresses the challenge of understanding RNA editing in noncoding regions, which constitute the majority of editing sites and whose functions are poorly understood.

Key Results

  • CircAI integrates data from 533 RNA-seq samples across eight species.
  • The database contains millions of RNA editing sites identified within circRNAs.
  • Detailed functional annotations are provided for each circRNA and its associated editing events, including predictions of miRNA binding changes, potential alterations in protein coding capacity, and structural stability.
  • The web interface allows for user-friendly search, browsing, visualization, and download of all data, facilitating efficient exploration.

Limitations and Future Work

  • Single-Cell Resolution: The authors acknowledge the need to incorporate single-cell RNA-seq data to achieve cellular-level resolution, which is currently challenging due to technical limitations in circRNA capture from poly(A)-selected libraries.
  • Drug Target Integration: Future work could involve incorporating data related to drug targets, particularly in the context of cancer chemoresistance, where circRNAs and RNA editing are increasingly implicated. This would enhance the database's utility for therapeutic research.
  • Species Expansion: The authors suggest expanding the database to include additional species, notably plant species, where RNA editing is also prevalent, to create a truly comprehensive resource across diverse organisms.
  • RBP Interactions: While miRNA interactions are covered, the paper mentions RNA-binding protein (RBP) interactions as a functional annotation, but the detailed methodology and results for RBP interactions are not explicitly detailed in the provided text. Future work could further elaborate on and expand these annotations.

Summary updated

BibTeX
@article{wang2025circai,
  author = {Yulan Wang and Lingxiao Zou and Jian Zhao and Jing Wu and Meng Zhang and Jingjing Liu and Quan Wang and Xuejiang Guo and Xiaofeng Song and Yixuan Wang},
  title = {{CircAI: a comprehensive database of CircRNA associated with A-to-I RNA editing}},
  journal = {Database J. Biol. Databases Curation},
  year = {2025},
  doi = {10.1093/database/baaf075},
}
Figure