MCLPMDA: A novel method for miRNA‐disease association prediction based on matrix completion and label propagation

Abstract

MiRNAs are a class of small non‐coding RNAs that are involved in the development and progression of various complex diseases. Great efforts have been made to discover potential associations between miRNAs and diseases recently. As experimental methods are in general expensive and time‐consuming, a large number of computational models have been developed to effectively predict reliable disease‐related miRNAs. However, the inherent noise and incompleteness in the existing biological datasets have inevitably limited the prediction accuracy of current computational models. To solve this issue, in this paper, we propose a novel method for miRNA‐disease association prediction based on matrix completion and label propagation. Specifically, our method first reconstructs a new miRNA/disease similarity matrix by matrix completion algorithm based on known experimentally verified miRNA‐disease associations and then utilizes the label propagation algorithm to reliably predict disease‐related miRNAs. As a result, MCLPMDA achieved comparable performance under different evaluation metrics and was capable of discovering greater number of true miRNA‐disease associations. Moreover, case study conducted on Breast Neoplasms further confirmed the prediction reliability of the proposed method. Taken together, the experimental results clearly demonstrated that MCLPMDA can serve as an effective and reliable tool for miRNA‐disease association prediction.

Adaptive multi-view multi-label learning for identifying disease-associated candidate miRNAs

“…and Yu et al . adopted matrix completion to recover the potential missing miRNA-disease associations[16, 17]. Zeng et al .…”

Benchmark of computational methods for predicting microRNA-disease associations

“…The network or graph algorithms focused on constructing miRNAs and/or disease similarity networks and efficient transferring miRNA-disease association labels between similar miRNAs and/or similar diseases in the network. Therefore, label propagation algorithm, which has the advantages of simplicity and efficiency on the miRNA/disease similarity networks, often constitutes the core component of the algorithm framework for this type of methods, e.g., MCLPMDA [15], LPLNS [16], SNMDA [17], and HLPMDA [18]. Nevertheless, more sophisticated algorithm designs are often crucial for successful prediction of miRNA-disease associations.…”

“…Nevertheless, more sophisticated algorithm designs are often crucial for successful prediction of miRNA-disease associations. For example, MCLPMDA employed matrix completion algorithm in addition to label propagation, LPLNS adopted linear neighborhood similarity when implementing label propagation, SNMDA introduced sparse neighborhood representation for building the similarity network, and HLPMDA took a heterogeneous label propagation approach to transfer association label among a heterogeneous set of similarity networks [15–18]. Other algorithms focusing on the specific topology of miRNA-disease association network have also been proposed, such as BNPMDA [19] that used the bipartite network projection and SACMDA [20] that made predictions with short acyclic connections in a heterogeneous graph.…”

A Novel Computational Model for Predicting microRNA–Disease Associations Based on Heterogeneous Graph Convolutional Networks

“…Over the past few years, numerous computational approaches have been built for predicting miRNA–disease associations [8,9,10,11,12,13,14,15,16], mainly in two categories: Similarity-based measure approaches and machine learning approaches. Nevertheless, several methods that use machine learning are essentially based on similarity measures and matrix factorization.…”

“…Global linear neighborhoods method (GLNMDA) reconstructs miRNA–disease similarity matrix, then implements label propagation to infer the latent interactions between miRNAs and diseases [10]. Yu et al [11] proposed a method for miRNA-disease association prediction based on Matrix completion and Label Propagation (MCLPMDA), which reconstructs the miRNA–disease similarity matrix using label propagation and matrix completion. Another structural perturbation method (SPM), which is also similarity-based link prediction method is applied to predict disease-related miRNAs [12].…”