Circular RNAs (circRNAs) are key regulators in the onset and progression of complex diseases, offering promise as diagnostic and prognostic biomarkers. However, most putative circRNA–disease associations remain experimentally unverified, largely due to the cost and time demands of wet-lab approaches. To bridge this gap, we present STRCDA (Sparse Topological Representation learning for CircRNA–Disease Associations). The pipeline first constructs fused similarity profiles for circRNAs and diseases by integrating diverse biological attributes. These initial matrices are then refined via random walk with restart to capture local features. Subsequently, a sparse-constrained dual-branch graph autoencoder extracts holistic topological embeddings from the refined local features and the known interaction network. Finally, an XGBoost classifier scores potential circRNA–disease pairs. On the CircR2Disease dataset, STRCDA achieves an AUC of 0.9771 and an AUPR of 0.9826 under five-fold cross-validation. Notably, 18 of the top 20 predicted associations were confirmed by independent experimental evidence, highlighting STRCDA’s efficacy as a robust tool for uncovering circRNA function in disease.
Circular RNAs (circRNAs) participate in gene regulation and disease progression, but experimental mapping of circRNA-disease associations (CDAs) remains costly and incomplete. We developed NARVGA, a hybrid prediction framework that combines non-negative matrix factorisation (NMF) with an adversarially regularised variational graph autoencoder (ARVGA). Functional, semantic and Gaussian interaction-profile kernel similarities are integrated; K-means-derived co-membership graphs reduce diffuse similarity connections; ARVGA learns nonlinear topological embeddings; and NMF captures complementary low-rank association patterns. An Extra Trees classifier then scores candidate circRNA-disease pairs. In the original transductive stratified five-fold benchmark on CircR2Disease, NARVGA achieved an area under the receiver operating characteristic curve (AUC) of 0.9868, an area under the precision–recall curve (AUPR) of 0.9887, 94.32% accuracy and a 94.44% F1-score. Ablation analysis identified cluster sparsification as the largest individual contributor, with additional gains from adversarial regularisation and low-rank augmentation. Without task-specific retuning, the model obtained AUCs of 0.9503 on LncRNADisease and 0.9700 on HMDDv4. Performance was stable across GCN depths and cluster settings, and 19 of the 20 highest-ranked hepatocellular carcinoma candidates had supporting published evidence. NARVGA provides a within-network prioritisation tool for RNA-disease association studies, although prospective validation, hard-negative assessment and independent-cohort testing remain necessary.