RESUMEN
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.
Asunto(s)
Neoplasias de la Mama/genética , Enfermedades Genéticas Congénitas/genética , Predisposición Genética a la Enfermedad , MicroARNs/genética , Algoritmos , Biología Computacional , Simulación por Computador , Femenino , Estudios de Asociación Genética , Enfermedades Genéticas Congénitas/epidemiología , HumanosRESUMEN
Recently, increasing studies have shown that miRNAs are involved in the development and progression of various complex diseases. Consequently, predicting potential miRNA-disease associations makes an important contribution to understanding the pathogenesis of diseases, developing new drugs as well as designing individualized diagnostic and therapeutic approaches for different human diseases. Nonetheless, the inherent noise and incompleteness in the existing biological datasets have 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 global linear neighborhoods (GLNMDA). Specifically, our method obtains a new miRNA/disease similarity matrix by linearly reconstructing each miRNA/disease according to the known experimentally verified miRNA-disease associations. We then adopt label propagation to infer the potential associations between miRNAs and diseases. As a result, GLNMDA achieved reliable performance in the frameworks of both local and global LOOCV (AUCs of 0.867 and 0.929, respectively) and 5-fold cross validation (average AUC of 0.926). Case studies on five common human diseases further confirmed the utility of our method in discovering latent miRNA-disease pairs. Taken together, GLNMDA could serve as a reliable computational tool for miRNA-disease association prediction.