Genomic fingerprints of the world's soil ecosystems

Abstract

Despite the explosion of soil metagenomic data, we lack a synthesized understanding of patterns in the distribution and functions of soil microorganisms. These patterns are critical to predictions of soil microbiome responses to climate change and resulting feedbacks that regulate greenhouse gas release from soils. To address this gap, we assay 1,512 manually curated soil metagenomes using complementary annotation databases, read-based taxonomy, and machine learning to extract multidimensional genomic fingerprints of global soil microbiomes. Our objective is to uncover novel biogeographical patterns of soil microbiomes across environmental factors and ecological biomes with high molecular resolution. We reveal shifts in the potential for (i) microbial nutrient acquisition across pH gradients; (ii) stress-, transport-, and redox-based processes across changes in soil bulk density; and (iii) greenhouse gas emissions across biomes. We also use an unsupervised approach to reveal a collection of soils with distinct genomic signatures, characterized by coordinated changes in soil organic carbon, nitrogen, and cation exchange capacity and in bulk density and clay content that may ultimately reflect soil environments with high microbial activity. Genomic fingerprints for these soils highlight the importance of resource scavenging, plant-microbe interactions, fungi, and heterotrophic metabolisms. Across all analyses, we observed phylogenetic coherence in soil microbiomes-more closely related microorganisms tended to move congruently in response to soil factors. Collectively, the genomic fingerprints uncovered here present a basis for global patterns in the microbial mechanisms underlying soil biogeochemistry and help beget tractable microbial reaction networks for incorporation into process-based models of soil carbon and nutrient cycling.IMPORTANCEWe address a critical gap in our understanding of soil microorganisms and their functions, which have a profound impact on our environment. We analyzed 1,512 global soils with advanced analytics to create detailed genetic profiles (fingerprints) of soil microbiomes. Our work reveals novel patterns in how microorganisms are distributed across different soil environments. For instance, we discovered shifts in microbial potential to acquire nutrients in relation to soil acidity, as well as changes in stress responses and potential greenhouse gas emissions linked to soil structure. We also identified soils with putative high activity that had unique genomic characteristics surrounding resource acquisition, plant-microbe interactions, and fungal activity. Finally, we observed that closely related microorganisms tend to respond in similar ways to changes in their surroundings. Our work is a significant step toward comprehending the intricate world of soil microorganisms and its role in the global climate.

Keywords: biogeochemistry; biogeography; carbon cycling; functional potential; machine learning; metaanalysis; metagenomics; nitrogen cycling; soil microbiology; soil microbiome.

Fig 1

Data set description. (A) Map of samples and distributions of (B) total assembled bp and (C) number of samples per biome according to Olson et al. (79). (D–I) Distribution of environmental variables across all soils from SoilGrids250m (80).

Fig 2

Highly abundant genomic attributes across all soils. Relative abundance of the top 15 most abundant soil microbial genomic attributes annotated by (A) KEGG orthology, protein family [(B) Pfam and (C) TIGRFAM], and (D) read-based taxonomy collated at the genus level.

Fig 3

Genomic fingerprints of high versus low pH and bulk density soils. Genomic attributes selected in at least one fingerprint for pH or bulk density are visualized. Pearson correlation coefficient is shown from blue to red in the primary heatmap, with bulk density correlations on the left and pH correlations on the right. The sidebars (respectively, from left to right) represent variable importance from random forest models of pH and bulk density and the type of attribute (e.g., KO, Pfam, and TIGRFAM). Please refer to Extended Data S1 for more information on the specific attributes associated with each soil environment. Read-based taxonomy collated at the genus level for each fingerprint is shown in Fig. 4.

Fig 4

Genomic fingerprints of soil biomes and phylogenetic distribution of selected soil taxa. (A) The normalized abundance of microbial attributes selected in at least one biome fingerprint is shown from blue to red. Variable importance is shown on the left side bar, and attribute type is shown on the right side bar. Please refer to Extended Data S1 for more information on the specific attributes associated with each soil environment. (B) Microbial genera selected in at least one fingerprint are depicted on a phylogenetic tree (generated by PhyloT). The innermost circle shows Pearson correlations with bulk density from green to purple and variable importance from green to orange. The middle circle shows Pearson correlations with pH from green to purple and variable importance from green to orange. The outer circle shows the normalized abundance of genera across biomes (from outer to inner: temperate broadleaf and mixed forests; tundra; Mediterranean forests, woodlands, and scrub; tropical and subtropical grasslands, savannas, and shrublands; temperate grasslands, savannas, and shrublands; boreal forests/taigas; deserts and xeric shrublands; temperate conifer forests; tropical and subtropical dry broadleaf forests; flooded grasslands and savannas; tropical and subtropical moist broadleaf forests; and montane grasslands and shrublands). Variable importance is shown in the outermost ring from green to orange.

Fig 5

Unsupervised clusters of microbial attributes associated with multiple soil factors. (A) KEGG orthology, (B) microbial genus, and (C) Pfam. TIGRFAM annotations did not yield a satisfactory model. Rows represent environmental variables. The total normalized abundance of an attribute per sample is plotted against each environmental variable. Lines and statistics represent linear regression.