Introduction to cncFinder

Traditionally, RNAs have been categorized as either coding and non-coding. Nevertheless, recent studies have demonstrated that certain non-coding RNAs also possess coding potential, known as coding and non-coding RNAs (cncRNAs), which exert pivotal regulatory functions across diverse biological processes. In this study, we present cncFinder, a novel model based on graph attention network, designed for the precise classification of bifunctional long non-coding RNAs. In the testing dataset, cncFinder achieved outstanding performance with an accuracy of 0.856, sensitivity of 0.851, specificity of 0.861, MCC of 0.712, and an AUC of 0.883, significantly outperforming existing models. Furthermore, the model demonstrated high accuracy and robustness on external testing and cross-species datasets, confirming its broad applicability. The datasets and source code are freely available at Here.