Calculation method for predicting interaction between circular RNA and micro RNA
By employing gated multi-head attention and graph structure learning, the accuracy and stability of circRNA–miRNA interaction prediction are improved, addressing the issues of insufficient semantic feature expression and robustness in existing models, and achieving efficient circRNA–miRNA interaction prediction.
Patent Information
- Application Number
- CN202511960057.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-24
- Publication Date
- 2026-03-31
AI Technical Summary
Existing circRNA–miRNA interaction prediction models suffer from insufficient semantic feature expression, inadequate utilization of structural information, and limited robustness and generalization ability, making it difficult to meet the needs of large-scale, systematic CMI mining.
We employ a gated multi-head attention and graph structure learning approach, combining graph attention convolution and a nonlinear classifier. By using the multi-head attention mechanism, we extract key semantic information, enhance feature representation capabilities, adapt to topological relationship modeling under conditions of feature noise and missing features, and improve model stability through an adaptive optimization algorithm.
This approach enables more reliable prediction of circRNA–miRNA interactions, improving prediction accuracy and model generalization ability while reducing experimental costs and time.