Refining Metabolic Network by Fuzzy Matching of Metabolite Names for Improving Metabolites Ranking Toward the Diseases

The aberrant form of the metabolites inside the human body is known as a disease in other terms. Metabolite’s impact on the complex diseases of humans is a very crucial one. Therefore, the depth study of the relationships between the metabolites and diseases is very beneficial in understanding pathogenesis. This chapter proposes a network-based method to identify and rank the disease-related metabolites. In this research, for calculating the disease similarity and the metabolite similarity, infer disease similarity and miRNA similarity have been applied. The miRNA-based disease-related metabolite identification performance was further improved by proposing the Subcellular Localization Weight Based miRNA Similarity (SLWBMISM), where subcellular localization of metabolites was considered to find the more similar metabolites. A fuzzy matching algorithm is adopted to identify the identical names from the two models so that the computational complexity of SLWBMISM can be reduced. After obtaining the identical names of metabolites, two models have been merged without duplication, and metabolite similarity is obtained using SLWBMISM. Then the process of reconstructing the metabolic network based on disease and metabolite similarity was done. At last, a random walk is executed on the reconstructed network to identify and rank disease-related metabolites. For this research work total of 1955 metabolites from network A, 883 metabolites from network B, and 662 diseases were extracted from the experimental datasets. Both networks are merged, and fuzzy matching of metabolite names is applied to avoid the redundant metabolite participating in the metabolite network more than one time. After applying the SLWBMISM method to the merged network, 594521 similarities have been obtained. The proposed method, FM-SLWBMISM, helps find more similar metabolites and enhances the efficiency of SLWBMISM in identifying metabolite prioritization toward complex diseases.