This work proposes MFCAMNet, a Multi-Feature fusion and Cross-Self-Attention model for MiRNA–Disease association prediction, which employs element-level addition, element-level multiplication, and concatenation operations to generate miRNA-disease pair features with rich information.
Abstract
Drive by the rapid envolution of deep-learning techniques, a large body of biological experiments bas has uncovered extensive associations between microRNAs (miRNAs) and complex human diseases, hig- hlighting the pivotal roles of miRNAs in pathogenesis. Elucidating these associations is essential for understanding disease mechanisms and developing preventive strategies. Traditional wet-lab validati- on, however, is notoriously labor- and resource-intensive, creating an urgent demand for efficient computational tools that can prioritize the most promising miRNA–disease candidates. Existing predictors predominantly rely on a single category of handcrafted features, thereby overlooking the complementary information embedded in multiple, heterogeneous data sources. Although a few recent attempts integrate diverse features, they usually exploit only a limited subset and fail to capture the intricate, non-linear relationships among them. To address these limitations, we propose MFCAMNet, a Multi-Feature fusion and Cross-Self-Attention model for MiRNA–Disease association prediction. Firstly, we construct multiple similarity matrices and employ two independent autoencoders with multi-source feature attention to obtain deep features of miRNA and disease to extract the inherent relationships between multiple features. Secondly, the proposed model employs element-level addition, element-level multiplication, and concatenation operations to generate miRNA-disease pair features with rich information. Finally, we use the encoder structure of the transformer to fuse the three deep features and predict all potential miRNA disease associations. We conducted comprehensive evaluations on the public HMDD v2.0 and HMDD v3.2 benchmark datasets. MFCAMNet achieved average AUCs of 0.9455 and 0.9420 under 5-fold and 10-fold cross-validation on HMDD v2.0, respectively, and an AUC of 0.9578 under 5-fold cross-validation on HMDD v3.2, outperforming state-of-the-art competitors. Case studies on breast, esophageal, and lung cancers further corroborate the reliability and practical utility of the proposed method.
GCN-XGB is a novel hybrid computational framework that integrates a two-layer Graph Convolutional Network with Extreme Gradient Boosting to improve the accuracy of miRNA-disease association prediction and is suggested to be a powerful and reliable tool for identifying potential disease-related miRNAs and prioritizing candidates for experimental validation.
Jie Zhou, Peishen Yan, Jia Qu et al.· Journal of Mechanics in Medi...· 0 citations
MRGBMDAT, a multi-relational graph encoder network with bilinear fusion for miRNA-disease association type prediction significantly outperforms five state-of-the-art methods across multiple evaluation metrics, exhibiting strong discriminative power and generalization capability.
Y. Sun, Wenjing Su, Si-Qi Zhu et al.· Bioinform.· 0 citations
A novel GNN framework, APKAGN, designed for predicting miRNA-disease associations significantly enhances the accuracy of MDAs prediction, offering a powerful tool for investigating disease mechanisms and identifying biomarkers.
Ru Nie, Yingkai Li, Zhengwei Li et al.· IEEE transactions on computa...· 0 citations
Abstract Motivation Circular RNAs (circRNAs) interact with microRNAs (miRNAs) to regulate gene expression and influence disease progression. However, traditional models tend to overlook the significant contributions of certain features when dealing with diverse sequence information, resulting in the inability to capture some deep topological structures and thus leaving room for improvement in prediction performance. Results We propose HMA-GCA, a novel framework that integrates hybrid manifold augmentation and gated cross-attention for CMI prediction. The model first constructs multi-scale descriptors by combining sequence-derived features (K-mer, CTD, Doc2Vec) and topological features (Role2Vec, node degree, neighborhood proximity). It then applies PCA for global linear projection and UMAP for local nonlinear manifold learning, enhancing feature representations while preserving intrinsic data geometry. A channel-wise gated cross-attention mechanism dynamically controls the injection of miRNA information into circRNA representations. Extensive experiments on three benchmark datasets show that HMA-GCA consistently outperforms state-of-the-art methods across multiple metrics. To ensure interpretability, we conducted SHAP analysis to quantify the contribution of each feature type, revealing that sequence-derived features and topological similarities are the most influential. Ablation studies confirm the necessity of each module, while case studies demonstrate that top-ranked predictions are supported by literature evidence. Overall, HMA-GCA not only achieves state-of-the-art predictive performance but also provides interpretable insights into the molecular features. Availability and implementation The source code and data are freely available at https://github.com/Lixunwind/Prediction-circ-mi-by-Gate.git. The implementation is based on Python and the required dependencies are listed in the repository.
Yun-peng Hu, Yansu Wang, Yifeng Bai et al.· Bioinform.· 0 citations
DC-MetaMG, a deep learning framework based on a causal disentanglement strategy that models association responses as the synergistic interplay between two complementary mechanisms: static binding and dynamic regulation, is proposed, demonstrating its suitability for training scenarios involving complex biological information.
Runzhou Tang, Xi Zhou, Yujie Qi et al.· IEEE journal of biomedical a...· 0 citations
Existing circRNA-miRNA interaction prediction methods have not fully exploited the spatial folding information in circRNAs and pre-miRNAs. Furthermore, current methods inadequately address intramolecular modal consistency and intermolecular discrimination across distinct modalities, leading to suboptimal discriminative performance. Traditional attention mechanisms are computationally intensive and lack fine-grained, balanced modal weight allocation, impeding efficient feature fusion. To address these limitations, we propose GSSCMI, a novel method comprising three key components. First, an information integration module incorporates similarity information, sequence features, and secondary structure features. Second, a multimodal contrastive learning module processes features across three modalities, enhancing both intra-modal consistency for individual circRNA/miRNA and inter-modal discrimination between different circRNAs/miRNAs. Third, we innovatively design an efficient co-attention mechanism that simultaneously modulates fused modalities through unified and fine-grained attention scores, achieving balanced fusion while significantly reducing computational overhead. Experimental results demonstrate GSSCMI outperforms existing methods, with improvements of 9.53% in MCC and 6.40% in F1. Ablation studies further show the efficient co-attention reduces convergence iterations by approximately 58%. Compared to the initial co-attention, it reduces computational cost by 42.94% while improving MCC by 13.32% and ACC by 6.69%. Additionally, we identified regulatory sites through structure-based A-to-I editing and elucidated sequence-level inter-token dependencies via attention visualization.
Lihao Sun, Xin Wang, Fang Wang et al.· IEEE journal of biomedical a...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
