Abstract
As is well known, soil is a complex ecosystem harboring the most prokaryotic biodiversity on the Earth. In recent years, the advent of high-throughput sequencing techniques has greatly facilitated the progress of soil ecological studies. However, how to effectively understand the underlying biological features of large-scale sequencing data is a new challenge. In the present study, we used 33 publicly available metagenomes from diverse soil sites (i.e. grassland, forest soil, desert, Arctic soil, and mangrove sediment) and integrated some state-of-the-art computational tools to explore the phylogenetic and functional characterizations of the microbial communities in soil. Microbial composition and metabolic potential in soils were comprehensively illustrated at the metagenomic level. A spectrum of metagenomic biomarkers containing 46 taxa and 33 metabolic modules were detected to be significantly differential that could be used as indicators to distinguish at least one of five soil communities. The co-occurrence associations between complex microbial compositions and functions were inferred by network-based approaches. Our results together with the established bioinformatic pipelines should provide a foundation for future research into the relation between soil biodiversity and ecosystem function.
Highlights
Soil is considered to be the most diverse natural environment on the Earth [1,2]
One of the most important issues in the field of soil ecology is to uncover the complex relationships between microbial compositions and functional diversity in soil
Proteobacteria was the most dominated phylum in the microbial community of soil, $70% abundance detected in all soil sites except for the microbiota in the desert samples (Figure 1A)
Summary
Soil is considered to be the most diverse natural environment on the Earth [1,2]. The soil microbial communities harbor thousands of different prokaryotic organisms that contain a substantial number of genetic information, ranging from 2,000 to 18,000 different genomes estimated in one gram of soil [3]. Based on traditional approaches for cultivating and isolating soil microorganisms, early studies have focused on culturable bacteria which only account for less than 1% of soil microbial populations [4]. These studies have already discovered many novel genes encoding interesting enzymes and antimicrobials in soils via functional screens and clone-based Sanger sequencing [1,5,6]. The majority of 16S rRNA gene-based studies are committed to the interpretation of community composition but poorly focus on the functional and metabolic properties in a microbial community [14]. The computational pipelines implemented by these prominent platforms are capable of processing many analysis tasks, but some approaches for special biological inference and graphical visualization still need to be complemented [18]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
