Computational Approaches to microRNA Disease Association Prediction

MicroRNAs are small noncoding RNAs that regulate gene expression after transcription and contribute to a wide range of biological processes. Changes in microRNA expression and activity have been linked to many human diseases, including cancer, cardiovascular disease, and neurodegenerative disorders. Identifying disease related microRNAs can help researchers understand disease mechanisms and prioritize candidates for biomarker development, diagnosis, and therapeutic research. However, experimental identification of these associations is often time consuming and difficult to perform on a large scale. Computational methods provide a useful way to combine existing biological knowledge with genomic, transcriptomic, and other molecular data to identify potential microRNA disease associations.Despite substantial progress in this field, several challenges remain. Available datasets are often incomplete and unevenly distributed across diseases. The selection of negative samples can affect model performance, while results obtained from a single dataset may not generalize to other diseases or independent datasets. In addition, many models provide limited insight into the biological basis of their predictions. This Research Topic aims to bring together studies that develop, evaluate, and apply computational methods for more reliable prediction of microRNA disease associations. We are particularly interested in work that connects model development with biological interpretation and independent validation.This Research Topic welcomes theoretical, methodological, and applied studies on computational prediction of microRNA disease associations. Subtopics of interest include:• Machine learning and deep learning methods for predicting microRNA disease associations.• Graph based models, heterogeneous network models, and network representation methods that integrate microRNAs, diseases, genes, proteins, and biological pathways.• Integration of genomic, transcriptomic, and other omics data to improve the identification and interpretation of microRNA disease associations.• Methods for addressing incomplete data, class imbalance, uncertain negative samples, and differences in data quality across diseases and datasets.• Model interpretation and uncertainty assessment to clarify the biological basis and reliability of computational predictions.• Transfer learning and cross disease prediction for diseases with limited known microRNA associations.• Systematic benchmarking of existing methods using independent datasets and consistent evaluation criteria.• Case studies that assess computational predictions using experimentally validated associations, independent patient cohorts, functional experiments, or other relevant biological evidence.We invite Original Research, Reviews, Systematic Reviews, Methods, and Perspectives. Manuscripts should provide a clear methodological contribution or offer meaningful biological insight. Studies based only on routine analysis of public databases, without methodological novelty, independent validation, or meaningful biological interpretation, are not the main focus of this Research Topic. Topic editors: Bo-Ya Ji, Lei Wang, Liangrui Pan, Leon Wong Source: https://www.frontiersin.org/research-topics/87387/computational-approaches-to-microrna-disease-association-prediction