Predict MiRNA-Disease Association with Collaborative Filtering

Abstract

The era of human brain science research is dawning. Researchers utilize the various multi-disciplinary knowledge to explore the human brain,such as physiology and bioinformatics. The emerging disease association prediction technology can speed up the study of diseases, so as to better understanding the structure and function of human body. There are increasing evidences that miRNA plays a significant role in nervous system development, adult function, plasticity, and vulnerability to neurological disease states. In this paper ,we proposed the novel improved collaborative filtering-based miRNA-disease association prediction (ICFMDA) approach. Known miRNA-disease associations can be viewed as a bipartite network between diseases and miRNAs. ICFMDA defined significance SIG between pairs of diseases or miRNAs to model the preference on the choices of other entities. The collaborative filtering algorithm is further improved by incorporating similarity matrices to enable the prediction for new miRNA or disease without known associations. Potential miRNA-disease associations are scored with the addition of bidirectional recommendation results with low computational cost. ICFMDA achieved a 0.9076 AUC of ROC curve in global leave-one-out cross validation, which outperformed the state-of-the-art models. ICFMDA is a compact and accurate tool for potential miRNA-disease association prediction. We hope that ICFMDA would be useful in future miRNA and brain researches,and achieve better understanding of the nervous system in molecular level, cellular level, cell change process, and thus can support the research of human brain.

Graph regularized L2,1-nonnegative matrix factorization for miRNA-disease association prediction

“…To evaluate our model's ability to predict disease-related miRNAs, we compared it with three state-of-art methods (ICFMDA [58], SACMDA [59] and IMCMDA [60]) by implementing two validation frameworks: global leave-one-out cross validation (global LOOCV) and fivefold cross validation (5-CV) according to the experimentally validated disease-related miRNAs in HMDD v2.0, Table 1 The effects of parameters α 1 and α 2 on the results of GRL 2, 1 -NMF γ 1 = 1,γ 2 = 0, θ 1 = 1,and θ 2 = 0 which gathered plenty of the known miRNA-disease associations [10].…”

MSCHLMDA: Multi-Similarity Based Combinative Hypergraph Learning for Predicting MiRNA-Disease Association

“…The prediction was considered false positive if the rank of the candidate sample was no lower than the threshold. The methods of EGBMMDA (Chen et al, 2018b), ICFMDA (Jiang et al, 2018), RLSMDA (Chen and Yan, 2014), and SACMDA (Shao et al, 2018) were implemented on the same dataset, and the parameters were set according to the values given in the original article. Finally, MSCHLMDA obtained the AUC of 0.9283 in LOOCV as shown in Figure 5 The AUCs of ICFMDA,EGBMMDA, SACMDA and RLSMDA in LOOCV are 0.9067, 0.9123, 0.8770, and 0.8426, respectively.…”

“…The first category is based on network analysis (Chen et al, 2012(Chen et al, , 2016Zeng et al, 2016Zeng et al, , 2018Li et al, 2017;Liu et al, 2017;Xiao et al, 2017;Zhong et al, 2017). Jiang et al (2018) designed the significance SIG of disease pairs or miRNA pairs and then developed a novel miRNA-disease association prediction (ICFMDA) method, which was used to improve the collaborative filtering approach. The collaborative filtering algorithm was further improved by incorporating similarity matrices to enable the prediction of a new miRNA and a particular disease without known associations.…”

Benchmark of computational methods for predicting microRNA-disease associations

“…For example, as the first model using decision tree learning, EGBMMDA has reported a global leave-one-out cross-validation (LOOCV) area under ROC curve (AUROC) greater than 0.9 [21]. And other machine learning algorithms, such as collaborative filtering adopted by ICFMDA [22] and latent feature extraction with positive samples taken by LFEMDA [23], also showed promising performances in cross-validation tests.…”