Related Experiment Video
Updated: May 11, 2025
Biotin-based Pulldown Assay to Validate mRNA Targets of Cellular miRNAs
Published on: June 12, 2018
Interval-Shared Information Integration and False-Negative Association Reduction in Multi-Source MiRNA-Disease
Abstract:
Numerous studies have demonstrated that microRNAs (miRNAs) play crucial roles in the development and progression of various diseases, making the identification of miRNA-disease association (MDA) essential for understanding human disease etiology. While several computational models have been developed to predict MDAs, challenges persist-particularly the limited consideration of information interactions among multi-source similarities and the presence of "false-negative" associations in the original topology. To address these issues, we propose ISFNMDA, a model designed to infer potential MDAs by leveraging multi-view collaborative learning for feature extraction and optimizing association topology through graph structure momentum contrastive learning. Specifically, multi-source similarities of miRNAs and diseases are mapped into a unified feature space via encoders. The Pearson correlation coefficient is employed to derive pairwise constraints between nodes, facilitating information interactions and constructing interval-shared information constraints. Subsequently, an inference graph learner models the representations to generate an inferred graph topology. By maximizing mutual information between the inferred topology and the original "false-negative" associations through momentum contrastive learning, the model effectively reduces spurious correlations. The final comprehensive representations and optimized graph structure are then used to predict potential MDAs. Experimental results demonstrate that ISFNMDA outperforms existing methods, and case studies further validate its predictive capability.
Insights
This study introduces ISFNMDA, a novel computational model for identifying microRNA-disease associations (MDAs). It enhances prediction accuracy by integrating multi-source similarities and optimizing graph topology, improving disease etiology understanding.
Area of Science:
- Bioinformatics
- Genomics
- Computational Biology
Background:
- MicroRNAs (miRNAs) are critical in disease development, making miRNA-disease association (MDA) identification vital for understanding disease etiology.
- Existing computational models for MDA prediction face challenges with multi-source similarity interactions and "false-negative" associations.
Purpose of the Study:
- To propose ISFNMDA, a novel computational model for inferring potential miRNA-disease associations (MDAs).
- To address limitations in current MDA prediction models, specifically regarding multi-source similarity integration and "false-negative" data.
Main Methods:
- Leveraging multi-view collaborative learning for feature extraction from multi-source miRNA and disease similarities.
- Optimizing association topology using graph structure momentum contrastive learning to reduce spurious correlations.
- Employing Pearson correlation coefficient for pairwise constraints and constructing interval-shared information constraints.
Main Results:
- ISFNMDA effectively integrates multi-source similarities into a unified feature space.
- The model optimizes graph topology by reducing spurious correlations through momentum contrastive learning.
- Experimental results show ISFNMDA outperforms existing methods in predicting potential MDAs, with case studies validating its efficacy.
Conclusions:
- ISFNMDA provides an effective computational framework for accurate miRNA-disease association prediction.
- The model's ability to handle complex data interactions and optimize graph structures offers significant advancements in disease etiology research.
- This approach enhances the understanding of disease mechanisms and facilitates the identification of novel therapeutic targets.
