Open AccessMicrobial Community Analysis based on Bipartite Graph Clustering of Metabolic Network
Author(s) -
Chen Zhang,
Li Deng
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1828/1/012092
Subject(s) - bipartite graph , cluster analysis , centrality , clustering coefficient , microbiome , microbial population biology , graph , gut microbiome , computer science , computational biology , mathematics , biology , bioinformatics , theoretical computer science , artificial intelligence , bacteria , genetics , combinatorics
Microbial metabolism network is significant for the study of the microbial community, which is crucial for microbiome related diseases, such as inflammatory bowel diseases (IBD). In order to understand the difference of gut microbial communities between IBD patients and healthy people. Firstly, metabolic bipartite networks—microbes-compound graph are proposed and then built for healthy people and IBD patients respectively, which preserve more metabolic information than the traditional unipartite network. Secondly, with the use of the community detection of LPA in weighted bipartite graphs, the community modules of the two networks are obtained. Finally, two networks are compared to analyse the differences between healthy people and IBD people from several perspectives, such as NMI, centrality, clustering coefficient, species and compounds related to IBD disease, and cross-validation is performed to prove that all results are reliable and robust. The result shows that the gut microbial communities of healthy people and IBD patients are quite different, and the diversity and stability declined. From the clustering results, it can be judged that the distribution of disease-related bacteria changed.