Pusan National University Researchers Develop AI Tool to Map MicroRNA Binding Sites on Circular RNAs

circRNA–microRNA interactions can be difficult to map precisely. Researchers developed an AI bioinformatics tool, circMAC, to predict candidate microRNA binding regions within full-length circular RNA isoforms at nucleotide resolution. By preserving circular sequence continuity and conditioning predictions on the specific microRNA, circMAC showed stronger performance than the evaluated RNA language models and rule-based screening approaches under the benchmark settings. The framework could help researchers prioritize candidate binding regions for experimental studies of gene regulation

BUSAN, South Korea, Oct. 22, 2026 /PRNewswire/ -- Circular RNAs (circRNAs) are loop-shaped molecules that can influence gene regulation through interactions with microRNAs (miRNAs). This matters because circRNAs can exist as different full-length isoforms, and their back-splice junction (BSJ) connects the two ends of the RNA. Treating a circular RNA like a linear transcript can therefore overlook sequence information near this junction.

A research team was led by Professor Giltae Song along with his colleagues from the School of Computer Science and Engineering at Pusan National University, South Korea, who developed circMAC, an AI bioinformatics tool for predicting candidate microRNA binding sites on circular RNAs. The framework takes a full-length circRNA isoform and a mature miRNA as inputs and predicts a binding probability for every circRNA nucleotide. This paper was made available online on 9 September 2026 and was published in Volume 27, Issue 5 of the Briefings in Bioinformatics on 9 September 2026.

"The key advance is moving beyond asking whether a circRNA and miRNA interact to asking where that interaction is likely to occur," said Prof. Song.

To preserve the distinctive circular sequence, circMAC combines three complementary sequence-modeling approaches. Attention captures information from across the circRNA, Mamba models sequential dependencies, and a one-dimensional convolutional neural network identifies short local sequence patterns associated with binding sites. Circular padding in the convolutional branch allows patterns spanning the BSJ to be considered continuously rather than interrupted at the sequence boundary. The model also uses cross-attention to incorporate information from the specific miRNA being analyzed, allowing each circRNA nucleotide to be evaluated in its miRNA-specific context. It also uses structure-aware pretraining based on base-pairing prediction.

The researchers evaluated circMAC using full-length circRNA isoforms supported by AGO-CLIP binding-site annotations. The main pair-disjoint dataset contained 19,674 training pairs and 7,838 test pairs, with 5.7% of nucleotide positions labeled as binding sites. At the 1000-nucleotide setting, circMAC with base-pairing prediction pretraining achieved an F1-macro score of 0.762, an AUROC of 0.902 and an AUPRC of 0.545, outperforming the evaluated pretrained RNA language models. It also showed the strongest overall performance among evaluated full-length models under isoform-disjoint and BSJ-disjoint tests.

"By retaining full-length sequence context and conditioning predictions on the specific miRNA, the framework can provide more focused candidate regions for follow-up experiments," said Prof. Song.

circMAC can help computational biology researchers identify potential microRNA binding regions on circular RNA at nucleotide resolution, enabling more targeted experimental validation. In an illustrative case, it concentrated prediction signals near an annotated BSJ region of circCDYL2. However, these predictions are intended to prioritize candidate binding regions rather than establish functional interactions or clinical applications.

Overall, circMAC provides a machine-learning framework for prioritizing candidate circRNA–miRNA binding regions. By combining RNA sequence modeling, circular sequence continuity, and miRNA-specific conditioning, it could support more precise studies of RNA-mediated regulation. Further validation using independent datasets and experiments is needed before broader biological or clinical applications can be considered.

Reference
Titles of original paper: circMAC: microRNA-conditioned binding-site localization on circular RNA isoforms
Journal: Briefings in Bioinformatics
DOI: https://doi.org/10.1093/bib/bbag492

About Pusan National University
Website: https://www.pusan.ac.kr/eng/Main.do

Media Contact
Goon-Soo Kim
82 51 510 7928
[email protected]

SOURCE Pusan National University