Search

The increase in biological data and the formation of various biomolecule interaction databases enable us to obtain diverse biological networks. These biological networks provide a wealth of raw materials for further understanding of biological systems, the discovery of complex diseases and the search for therapeutic drugs. However, the increase in data also increases the difficulty of biological networks analysis. Therefore, algorithms that can handle large, heterogeneous and complex data are needed to better analyze the data of these network structures and mine their useful information. Deep learning is a branch of machine learning that extracts more abstract features from a larger set of training data. Through the establishment of an artificial neural network with a network hierarchy structure, deep learning can extract and screen the input information layer by layer and has representation learning ability. The improved deep learning algorithm can be used to process complex and heterogeneous graph data structures and is increasingly being applied to the mining of network data information. In this paper, we first introduce the used network data deep learning models. After words, we summarize the application of deep learning on biological networks. Finally, we discuss the future development prospects of this field. [ABSTRACT FROM AUTHOR]

Disease–disease relationships (e.g., disease comorbidities) play crucial roles in pathobiological manifestations of diseases and personalized approaches to managing those conditions. In this study, we develop a network-based methodology, termed meta-path-based Disease Network (mpDisNet) capturing algorithm, to infer disease–disease relationships by assembling four biological networks: disease–miRNA, miRNA–gene, disease–gene, and the human protein–protein interactome. mpDisNet is a meta-path-based random walk to reconstruct the heterogeneous neighbors of a given node. mpDisNet uses a heterogeneous skip-gram model to solve the network representation of the nodes. We find that mpDisNet reveals high performance in inferring clinically reported disease–disease relationships, outperforming that of traditional gene/miRNA-overlap approaches. In addition, mpDisNet identifies network-based comorbidities for pulmonary diseases driven by underlying miRNA-mediated pathobiological pathways (i.e., hsa-let-7a- or hsa-let-7b-mediated airway epithelial apoptosis and pro-inflammatory cytokine pathways) as derived from the human interactome network analysis. The mpDisNet offers a powerful tool for network-based identification of disease–disease relationships with miRNA-mediated pathobiological pathways. Disease comorbidity analysis from microRNA regulatory networks: Identification of disease–disease relationships play essential roles in development of personalized approaches to managing those conditions. A team led by Dr. Feixiong Cheng at Cleveland Clinic developed a network-based methodology (termed mpDisNet) to infer clinically relevant disease–disease relationships from miRNA regulatory networks. mpDisNet shows high performance in identification of clinically reported disease–disease relationships by a unique integration of miRNA–gene–disease networks under the human interactome model. Importantly, mpDisNet successfully identifies a network-based relationship between asthma and chronic obstructive pulmonary disease driven by underlying miRNA-mediated pathobiological pathways. From a translational perspective, if broadly applied, mpDisNet would offer a powerful network-based tool for understanding of clinical comorbidities for multiple complex diseases from heterogeneous biological networks, a significant challenge of precision medicine. [ABSTRACT FROM AUTHOR]