Agricultural practices affect soil microbes which are critical to soil health and sustainable agriculture. To understand prokaryotic and fungal assembly under agricultural practices, we use machine learning-based methods. We show that fertility source is the most pronounced factor for microbial assembly especially for fungi, and its effect decreases with soil depths. Fertility source also shapes microbial co-occurrence patterns revealed by machine learning, leading to fungi-dominated modules sensitive to fertility down to 30 cm depth. Tillage affects soil microbiomes at 0-20 cm depth, enhancing dispersal and stochastic processes but potentially jeopardizing microbial interactions. Cover crop effects are less pronounced and lack depth-dependent patterns. Machine learning reveals that the impact of agricultural practices on microbial communities is multifaceted and highlights the role of fertility source over the soil depth. Machine learning overcomes the linear limitations of traditional methods and offers enhanced insights into the mechanisms underlying microbial assembly and distributions in agriculture soils.

Background: Annually reoccurring microbial populations with strong spatial and temporal variations have been identified in estuarine environments, especially in those with long residence time such as the Chesapeake Bay (CB). However, it is unclear how microbial taxa cooccurr and how the inter-taxa networks respond to the strong environmental gradients in the estuaries. Results: Here, we constructed co-occurrence networks on prokaryotic microbial communities in the CB, which included seasonal samples from seven spatial stations along the salinity gradients for three consecutive years. Our results showed that spatiotemporal variations of planktonic microbiomes promoted differentiations of the characteristics and stability of prokaryotic microbial networks in the CB estuary. Prokaryotic microbial networks exhibited a clear seasonal pattern where microbes were more closely connected during warm season compared to the associations during cold season. In addition, microbial networks were more stable in the lower Bay (ocean side) than those in the upper Bay (freshwater side). Multivariate regression tree (MRT) analysis and piecewise structural equation modeling (SEM) indicated that temperature, salinity and total suspended substances along with nutrient availability, particulate carbon and Chl a, affected the distribution and co-occurrence of microbial groups, such as Actinobacteria, Bacteroidetes, Cyanobacteria, Planctomycetes, Proteobacteria, and Verrucomicrobia. Interestingly, compared to the abundant groups (such as SAR11, Saprospiraceae and Actinomarinaceae), the rare taxa including OM60 (NOR5) clade (Gammaproteobacteria), Micrococcales (Actinobacteria), and NS11-12 marine group (Bacteroidetes) contributed greatly to the stability of microbial co-occurrence in the Bay. Modularity and cluster structures of microbial networks varied spatiotemporally, which provided valuable insights into the ‘small world’ (a group of more interconnected species), network stability, and habitat partitioning/preferences. Conclusion: Our results shed light on how estuarine gradients alter the spatiotemporal variations of prokaryotic microbial networks in the estuarine ecosystem, as well as their adaptability to environmental disturbances and co-occurrence network complexity and stability.

Chesapeake Bay is a large, dynamic, and productive temperate estuary, which plays an important role in transporting terrestrial materials into the Atlantic Ocean. The bacterioplankton communities experience strong environmental changes both seasonally and along the central axis of the Bay. The goal of this study was to explore the spatiotemporal and interannual patterns of the bacterioplankton communities in the estuary using high‐throughput sequencing of 16S rRNA genes. Deep sequencing uncovered many new bacterial groups not found in earlier studies, and showed that Chesapeake Bay contained highly diverse and dynamic bacterial communities. Bacterioplankton in the Bay in summer and autumn were more diverse and stable than those in winter and spring. The spatial pattern was overruled by strong temporal variations when the samples from all seasons were analyzed together. But, the spatial trend became more visible when samples were analyzed separately in individual seasons. Bacterial diversity indices decreased from the upper Bay to lower Bay. The spatial and temporal distributions of bacterioplankton were best explained by variations of salinity and temperature. In addition, nutrient availability (nitrogen and phosphorus), chlorophyll a, and light availability also contributed significantly to the spatiotemporal variation. Despite the interannual variation of the bacterioplankton communities, a stable and reoccurring spatiotemporal pattern was observed in the Chesapeake Bay, suggesting limited functional redundancy for planktonic estuarine microbiomes. Our results highlight the resilience of the planktonic bacterial communities in estuaries with long residence time, and different bacterial groups have unique responses/adaptation to dynamic environmental gradients.

Dissolved organic matter (DOM) serves as a major carbon and nutrient pool in oceans, and recalcitrant DOM are the primary sources for carbon sequestration in depths. Here, we demonstrate the critical roles of algae-associated microorganisms (mainly heterotrophic bacteria) in the transformation of recalcitrant dissolved organic matter through laboratory cultures of a model diatom, Skeletonema dohrnii .

In addition to inhabiting extreme territories, Archaea are widely distributed in common environments spanning from terrestrial to aquatic environments. This study investigated and compared archaeal community structures from three different habitats (representing distinct environments): agriculture soils (from farming system trials FST, PA, United States), freshwater biofilms (from White Clay Creek, PA, United States), and estuary water (Chesapeake Bay, United States). High-throughput sequencing of 16S rRNA genes indicated that Thaumarchaeota, Euryarchaeota, Nanoarchaeota, Crenarchaeota, and Diapherotrites were the commonly found dominant phyla across these three environments. Similar to Bacteria, distinct community structure and distribution patterns for Archaea were observed in soils vs. freshwater vs. estuary. However, the abundance, richness, evenness, and diversity of archaeal communities were significantly greater in soils than it was in freshwater and estuarine environments. Indicator species (or amplicon sequence variants, ASVs) were identified from different nitrogen and carbon cycling archaeal groups in soils (Nitrososphaerales, Nitrosotaleales, Nitrosopumilales, Methanomassiliicoccales, Lainarchaeales), freshwater biofilms (Methanobacteria, Nitrososphaerales) and Chesapeake Bay (Marine Group II, Nitrosopumilales), suggesting the habitat-specificity of their biogeochemical contributions to different environments. Distinct functional aspects of Archaea were also confirmed by functional predictions (PICRUSt2 analysis). Further, co-occurrence network analysis indicated that only soil Archaea formed stable modules. Keystone species (ASVs) were identified mainly from Methanomassiliicoccales, Nitrososphaerales, Nitrosopumilales. Overall, these results indicate a strong habitat-dependent distribution of Archaea and their functional partitions within the local environments.

Phytoplankton are the basis of primary production and play important roles in regulating energy export in marine ecosystems. Compared to other regions, chromophytic phytoplankton is considerably understudied in the Bay of Bengal (BOB). Here we investigated community structure and spatial distribution of chromophytic phytoplankton in the BOB by using RuBisCO genes (Form ID rbcL). High throughput sequencing of rbcL genes revealed that Diatoms, Cyanobacteria (Cyanophyceae), Pelagophyceae, Haptophyceae, Chrysophyceae, Eustigamatophyceae, Xanthophyceae, Cryptophyceae, Dictyochophyceae, and Pinguiophyceae were the most abundant groups recovered in the BOB. Abundances and distribution of Bacillariophyceae and Pelagophyceae were further verified using quantitative PCR analyses which showed the dominance of these groups near the Equator region (p < 0.01) where upwelling was likely the source of nutrients. Further, redundancy analysis demonstrated that temperature was an important environmental driver in structuring distributions of Cyanophyceae and dominant chromophytic phytoplankton. Morphological identification and quantification confirmed the dominance of Diatoms, and also detected other cyanobacteria and dinoflagellates that were missing in our molecular characterizations. Pearson correlations of these morphologically identified phytoplankton with environmental gradients also indicated that nutrients and temperature were key variables shaping community structure. Combination of molecular characterization and morphological identification provided a comprehensive overview of chromophytic phytoplankton. This is the first molecular study of chromophytic phytoplankton accomplished in the BOB, and our results highlight a combination of molecular analysis targeting rbcL genes and microscopic detection in examining phytoplankton composition and diversity.