Methanogenesis has been widely accepted as an ancient metabolism, but the precise evolutionary trajectory remains hotly debated. Disparate theories exist regarding its emergence time, ancestral form, and relationship with homologous metabolisms. Here, we report the phylogenies of anabolism-involved proteins responsible for cofactor biosynthesis, providing new evidence for the antiquity of methanogenesis. Revisiting the phylogenies of key catabolism-involved proteins further suggests that the last Archaeacommon ancestor (LACA) was capable of versatile H₂-, CO₂-, and methanol-utilizing methanogenesis. Based on phylogenetic analyses of the methyl/alkyl-S-CoM reductase family, we propose that, in contrast to current paradigms, substrate-specific functions emerged through parallel evolution traced back to a non-specific ancestor, which likely originated from protein-free reactions as predicted from autocatalytic experiments using cofactor F₄₃₀. After LACA, inheritance/loss/innovation centred around methanogenic lithoautotrophy coincided with ancient lifestyle divergence, which is clearly reflected by genomically predicted physiologies of extant archaea. Thus, methanogenesis is not only a hallmark metabolism of Archaea, but the key to resolve the enigmatic lifestyle that ancestral archaea took and the transition that led to physiologies prominent today.

We recently reported on the co-occurrence and cooperation of aerobic and anaerobic nitrifying bacteria in a full-scale attached growth systems for nitrogen removal from wastewater. While canonical ammonia oxidizing bacteria were present, aerobic ammonia oxidation was primarily driven by complete ammonia oxidizing (comammox) bacteria within the Candidatus Nitrospira nitrosa cluster. Further, Brocadia-like anaerobic ammonia oxidizing (anammox) bacteria leverage the lack of oxygen in dense attached growth biofilms to remove nitrite and residual ammonia further driving nitrogen loss. While aerobic and anaerobic ammonia removal was measured in the full-scale reactors and in batch assays, the precise mechanism for nitrite availability for anammox bacteria is unclear. Specifically, nitrite for anaerobic ammonia oxidation could either be made available by (1) partial nitrification of ammonia by comammox bacteria or canonical ammonia oxidizing bacteria, (2) partial denitrification of nitrate by denitrifying bacteria, or (3) nitrite comproprotionation by comammox or other novel bacteria. Identifying the precise mechanism for nitrite availability is crucial for optimizing nitrogen loss via anaerobic ammonia removal. To better understand microbial processes governing nitrite production, we investigated the changes in transcriptional profiles of the microbial community in this attached growth system across several dissolved oxygen setpoints. We quantified changes in the expression of nitrifying bacteria, and changes in the bulk community's protein synthesis potential by performing RT-qPCR and amplicon sequencing of genes involved in all known aerobic and anaerobic biotransformation’s of ammonia, nitrite, and nitrate. These analyses will be complemented with meta-transcriptomics to further determine pathways regulating nitrite availability.

Rising global populations have amplified food scarcity and ushered in the development of genetically modified (GM) crops to overcome these challenges. Modern GM crops that use small interference RNA (siRNA) to control gene expression have become increasingly common in the United States, but remain controversial due to the uncertainty regarding the unintended release of its genetic material into the environment and possible downstream effects on environmental health. The release of siRNA from crop tissues during cultivation or decomposition can contribute to increased persistence and bioavailability and possible uptake by soil microbial communities. This may lead to non-target gene silencing and the disruption of microbial ecosystem services such as nutrient cycling and soil fertility. Further information on the environmental effects of these genetic constructs is required as well as research focused on the genetic transfer to belowground microbes, transport, and fate. Thus, the goal of this research is to determine the ecological impact of siRNAs by evaluating their influence on the growth and functions of soil microbial communities. Preliminary results have shown a statistically significant difference in growth patterns when model microorganisms are exposed to siRNAs in a stressful environment. In the tested microcosms, we observed significant decreases in carrying capacity, growth rate, and doubling time (p< 0.05 were considered statistically significant), suggesting certain environmental conditions facilitate horizontal gene transfer of the siRNAs through microbial transformation. Using a GeoChip array, we aim to identify the specific genes that are altered and the consequences this may have for microbiome health and resilience.

Activated sludge (AS) systems are widely used for wastewater treatment, but their performance is often difficult to predict due to the complex microbial community involved. Herein, a hybrid model, integrating Bayesian Network (BN), Artificial Neural Network (ANN), and Activated Sludge Model No.1 (ASM1), was developed to simulate the performance of AS based on the prediction of kinetic parameters of the microbial community. To train the model, we collected 318 samples from previous publications, which were run under different operating conditions, and performed 16S rRNA amplicon sequencing to characterize the microbial community. Historical populations and time spans were defined as nodes in the BN to determine the input nodes of a subsequent artificial neural network (ANN). Different numbers of populations were selected based on their microbial abundance and occurrence to explore the ANN with the optimal number of populations. The populations with 0.5% of abundance and 30% of frequency were found to best fit the ANN. The Bray-Curtis similarity between the predicted and observed microbial community structure reached 92% at the family level. The model could further infer microbial kinetic parameters, which were then fed into the ASM1 to simulate AS performance. Iteration of this hybrid model would provide near-real-time feedback on AS performance. The hybrid model provides a promising approach for optimizing the operation and maintenance of AS systems for efficient wastewater treatment.

To respond to emerging viral threats and accurately assign virus removal credits in treatment processes, researchers must develop models to predict kinetics for virus-disinfectant reactions. These models rely on accurate, high-quality, and complete datasets collected through literature review and targeted experimentation. We will discuss predictive and statistical models of free chlorine inactivation of viruses. We used data from a systematic review to create an optimized linear mixed model to estimate chlorine inactivation rate constants. The model contained experimental variables, including temperature and pH, virus-level variables, including name and Baltimore class, feature interactions, and a paper identifier to account for correlated errors within papers. Comparisons of model coefficients generate insight into the drivers of inactivation; for example, all experimental conditions included in the model produced estimates that differed from reference conditions with p < 0.01, but only 12 of the 27 included virus species had estimates that differed from MS2 coliphage with p < .05. We then used this model to estimate rate constants for all viruses in the dataset at uniform conditions and used machine learning to predict inactivation rate from virus characteristics. Together, the models generate insight into the relative importance of virus and experimental characteristics, allow estimation of virus inactivation with novel viruses and in novel virus-environment combinations, and motivate targeted laboratory studies. This work will be discussed in the context of past modeling work on UV inactivation of viruses and a current systematic review of chlorine dioxide, chloramine, and peracetic acid disinfection as a precursor to modeling,