The human microbiome plays important roles in human health and disease. Previous microbiome studies focused mainly on single pure species function and overlooked the interactions in the complex communities on system-level. A metagenomic approach introduced recently integrates metagenomic data with community-level metabolic network modeling, but no comprehensive tool was available for such kind of approaches. To facilitate these kinds of studies, we developed an R package,
The human microbiome has been proved to play a key role in human health and disease. Various microorganisms species along with wide range of interactions among them structure the microbial communities as inherently complex ecosystems across different human body sites, such as gut, oral cavity, and skin [
Traditional culture-dependent methods are restricted by the small number of cultured species and often fail to describe the less abundant species [
A key challenge of applying metagenomics to microbial community is metabolic network reconstruction from metagenomic data. Previous studies focused mainly on the “parts list” of the microbiome and overlooked the interactions in the complex communities on system-level [
To examine whether enzymes that are associated with a specific host state exhibit some topological features in the SSN, we calculated most common topological properties (Table
The topological properties calculated in the SSN.
Topological features | Description |
---|---|
Betweenness centrality | The fraction of shortest paths between node pairs that pass through the node |
|
|
Clustering coefficient | The number of triangles (3 loops) that pass through this node |
|
|
PageRank | The number and PageRank metric of all nodes that link to the node |
|
|
Degree | The number of edges connected to the node |
A typical analysis pipeline for metagenomic systems biology supported by this package (Figure
A typical analysis pipeline supported by the mmnet package.
More detailed description of these functions and the package instructions is referred to in reference manual
To illustrate the analysis pipeline made by this package, we use a part of the public dataset containing 18 microbiomes from 6 obese and lean monozygotic twin pairs and their mothers [
The relative abundances of enzymatic genes in the two samples were estimated from the functional annotations, and then the corresponding SSNs were built. For these two samples, 1345 KOs were identified in total. The correlation coefficient tested with Pearson’s method is 0.92, which indicated that the relative enzymatic gene abundance across these two samples was highly concordant:
Based on the SSNs, we intuitively explored the correlations between the topological features of enzyme in the SSN and their abundance (Figure
Notably, only two samples were taken for testing
The metagenomic approach on metabolic network provides a system-level understanding of the microbiome and gains insight into variation in metabolic capacity. It is very useful for studying the metabolic activity and specifically complex inherent interactions by serving the microbial community as a single supraorganism. In this paper, we present the
project name: mmnet; project home page: operating system(s): platform independent; programming language: R; other requirements: R 3.1.0 or higher; license: GNU GPLv2; any restrictions to use by nonacademics: none.
The authors declare that they have no conflict of interests.
Yang Cao implemented the package and wrote the user manual. Fei Li designed the structure and interface of the software and drafted the paper. Xiaofei Zheng, Xiaochen Bo participated in the design of the package and helped to draft the paper. All authors read and approved the final paper.
National Major Science and Technology Special Projects for New Drugs (2013ZX09304101), National Major Science and Technology Special Projects for Infectious Diseases (2013ZX10004216), National Key Technologies R&D Program for New Drugs (2012ZX09301-003), National Science & Technology Pillar Program of China (2012BAI29B07), and National Nature Science Foundation of China (81102419). The authors thank Dr. Folker Meyer for kind response about the usages of MG-RAST API.