RMH

Cellular selection has emerged as a powerful approach for high throughput de novo discovery of bioactive cyclic peptides (CPs). This strategy coexpresses genetically encoded CP libraries with the target of interest in cells and uses customizable selection systems to link desired CP activities to a selectable phenotype to directly identify functional inhibitors. Cellular selections have yielded many CP inhibitors against a diverse range of therapeutic targets and have been continuously empowered by expansions in the CP diversity and selection repertoire. In this review, we summarize studies on cellular selection using genetically encoded CP libraries and discuss opportunities for future application of this technology.

Efficient transcription factor (TF) screening is essential for the rational optimization of industrial microbial strains. Here, we present an absorbance-activated laccase-based biosensor (AALB) platform designed for rapid functional screening of TFs in Trichoderma reesei. By using laccase as a reporter, the AALB platform offers a robust and cost-effective screening system for TF functional evaluation. The secreted heterologous laccase also complements T. reesei’s native cellulolytic machinery, expanding its enzymatic repertoire for lignocellulosic biomass deconstruction. The core of the AALB platform is a Pcbh1-regulated laccase expression cassette, enabling sensitive and efficient selection of TF variants based on absorbance readouts. To further improve sensitivity, we introduced a quadruple repeat of the cis-element-dense segment in the cbh1 promoter and integrated a high-activity heterologous laccase at the SLP1 site of the T. reesei genome. Following a miniaturized 2 day pre-culture in 24-well plates, the AALB platform enables TF variant ranking within 30 min of absorbance readout in 200 μL 96-well reactions, reducing the need for extended shake-flask fermentation and separate downstream enzymatic or molecular assays during the initial screen. Using AALB, we successfully identified TF variants with robust transcriptional regulation. The screening results were validated through transcript quantification of laccase and XYR1 and extracellular laccase activity assays. Notably, the XYR1 mutation XYR1_V821F, selected via AALB, significantly increased laccase transcript levels compared to PoLAC-SLP1 chassis control, with laccase activity reaching 17.74 ± 0.76 IU/mLa 2.7-fold improvement. Further characterization of the AALB-selected XYR1_V821F strain showed increased extracellular laccase activity together with increased xylanase and filter paper activities, with extracellular xylanase and filter paper activities increased by 2.7- and 1.9-fold, respectively, and enhanced rice straw hydrolysis under the tested fermentation and hydrolysis conditions. The AALB platform provides valuable insights into the synthetic biology and metabolic engineering of T. reesei, facilitating the development of industrial strains for lignocellulosic biomass processing and enzyme production.

The recombinant expression of integral membrane proteins is notoriously challenging. One way to address this challenge is via computational genotype-to-phenotype models that determine how particular sequence features correlate with protein expression levels. However, the potential of such approaches is yet to be fully realized, at least partly because so few expression datasets are available. Here, we study the sequence-to-expression relationships of a variant library originally derived from combinatorial computational design. The controlled sequence diversity of this library makes this new dataset directly compatible with lightweight off-the-shelf bioinformatic tools. The expression phenotype of the entire library was first assessed in the widely used recombinant host E. coli. We selected a relatively small and balanced dataset of 2055 protein sequences assigned to binary classes of either “high” or “low” expression and used these sequences to train a sequence-to-expression classifier using supervised machine learning. This trained model was then used to infer the expression of >10,000 unmeasured sequences, and experimental validation of these predictions for 12 test variants achieved a 100% success rate. Using tools from explainable AI, we identified specific sequence positions and substitutions that are most important in dictating cellular expression levels. This analysis was validated by model-guided protein engineering that achieved an 8-fold increase in the purification yield of a poorly expressing variant. Our results show that computational protein design in tandem with supervised learning leads to effective models for the discovery of protein variants with improved expression phenotypes and can decode the molecular basis of membrane protein expression.

Bacteriophages represent a largely untapped reservoir of regulatory elements that can expand the genetic toolbox for engineering non-model hosts. Using ONT-cappable-sequencing, we mapped full-length primary transcripts across five Pseudomonas phages and identified 232 transcription start sites (TSSs) and 176 transcription termination sites (TTSs), including 38 intrinsic factor-independent terminators. Motif discovery revealed 59 σ70-like promoters in KIL5, 37 in KIL3b, and unique phage-encoded RNA polymerase (RNAP) promoter motifs for KNP, ϕ2, and pphageB21, enabling discrimination between host- and phage-dependent transcriptional control. To validate these identifications, a subset of ten promoters and five terminators was quantitatively characterized in vivo in Pseudomonas fluorescens strain GL-S-306, showing promoter activities spanning a large dynamic range and terminator efficiencies allowing complete transcriptional termination. These data establish the first experimentally validated, phage-derived library of promoters and terminators for P. fluorescens and Pseudomonas syringae, providing chassis-specific regulatory parts that enable more finetuned gene-expression than current cross-species SynBio tools. Our results position phage transcriptomics as a scalable strategy for mining regulatory elements tailored to non-model bacterial hosts, while also supporting phage design and engineering efforts and accelerating the development of new synthetic biology platforms.

Data curation-the process of collecting, cleaning, and joining raw datasets for unified analysis-is a time-consuming yet crucial part of any data science project. With the emergence of powerful agentic coding tools, it is tempting to "vibe curate"-i.e., to prompt AI agents to perform data curation operations and blindly trust their outputs in order to finish work more quickly. However, this can exacerbate two key challenges that already plague manual data curation workflows: reproducibility-the ability to trace the exact set of modifications applied to raw datasets-and verifiability-the ability to audit changes and confirm that data is processed properly. To address these issues, we introduce Scrub Data, a framework that enables the use of AI agents in data curation workflows while transparently maintaining both reproducibility and verifiability. The framework involves an interactive data curation loop between a user and an AI agent, backed by a data provenance graph that tracks and versions all dataset updates. The data graph requires that all updates are formatted as self-contained executable steps, allowing any version of a dataset to be reconstructed from raw data by playing forward the transformations stored in the graph. This workflow takes place within a lightweight web application that is designed to be modified by users and agents in real-time to create custom visualization tools for identifying data curation needs and verifying outcomes. We demonstrate the usefulness of the framework with a real data curation use case from animal movement ecology. Starting with 89M raw GPS coordinates, we use the framework to curate a benchmark dataset of 2.5M coordinates to be used for machine learning or statistical analysis. The framework is available for use and extension at https://github.com/justinkay/scrubdata.

Microbial genomes encode compact molecular machines and diverse non-coding RNAs (ncRNAs) essential to gene regulation, pathogenesis, and many foundational biotechnologies. However, annotation remains largely protein-centric and homology-driven. Here, we introduce Minerva, a framework for coevolutionary mining that uses genome language models to predict local interactions directly from sequence as two-dimensional maps. Introducing two complementary techniques, categorical Jacobian fingerprinting and interaction heads, we demonstrate fast, accurate, alignment-free prediction of ncRNA base-pairing, monomeric protein contacts, and repetitive sequence motifs. Applied to 150 bacterial genomes, Minerva recovers known systems and predicts that 84.3% of predicted intergenic base-pairing falls outside of known annotations. In Pseudomonas, we find that the widespread TwoAYGGAY ncRNA family carries large secondary-structure extensions and is often flanked by short upstream repetitive motifs and larger downstream genomic repeats. Interpreting the coevolution maps, we find that Minerva emergently detects open-reading-frame (ORF) signatures at the DNA level despite never being trained to do so. In prophages within these genomes, we discover that Unknown Group 27 (UG27) reverse transcriptase systems encode arrays of structurally conserved yet sequence-diverse ncRNAs that template complementary DNA (cDNA) hairpin products. Together, these results establish coevolutionary mining as a scalable route to genome annotation and biological discovery across the rapidly expanding microbial universe.

Electroactive microbiomes are mixed microbial communities in which one or more members exchange electrons with extracellular minerals, redox-active compounds, electrodes, or partner organisms and in which community interactions materially influence net electron flow. This review develops a mechanism–interface–function–readiness framework that connects extracellular electron transfer (EET) with biofilm ecology, microbiome–electrode organization, environmental process performance, and translational maturity. Direct transfer through multiheme cytochromes and conductive structures, mediated transfer through soluble redox shuttles, and interspecies electron transfer are evaluated as system-dependent pathways rather than universally ranked mechanisms. Particular emphasis is placed on wastewater treatment and resource recovery, anaerobic digestion, pollutant and metal transformation, soil and sediment bioelectrochemistry, carbon conversion, microbial fuel cells, microbial electrolysis cells, electro-fermentation, and microbial electrosynthesis. Multi-omics, metabolic modelling, synthetic biology, advanced materials, and artificial intelligence are examined according to the strength of their direct evidence in electroactive systems. Across applications, performance depends strongly on reactor configuration, electrode properties, inoculum, biofilm architecture, mass and charge transport, substrate loading, and normalisation basis, making unqualified cross-study numerical comparison inappropriate. Major barriers are long-term stability, mechanistic attribution in mixed communities, standardisation, scale-up, energy and mass balances, biosafety, techno-economic feasibility, and regulatory compatibility. mfc

The apoplast is a primary location for interactions between plants and invasive pathogens and is the site where many pathogen-secreted effectors are perceived by cell surface immune receptors to activate defence responses. However, our understanding of apoplastic interactions remains limited because protein interactions are difficult to investigate in this harsh extracellular environment. TurboID-based proximity labelling (PL) has emerged as a powerful approach for studying protein interactions in plants, though its application has to date been restricted to intracellular proteins. Here, we designed and validated a TurboID-based PL strategy for investigating protein interactions in the leaf apoplast using the well-characterised interaction between the Phytophthora infestans elicitor INF1 and the receptor-like protein (RLP) REL in Nicotiana benthamiana. Transient expression of SP-INF1-TurboID (INF1-T) induced a cell death (CD) response comparable to that triggered by native INF1, demonstrating that fusion of the TurboID tag did not impair INF1 recognition by REL. Apoplastic localisation of both INF1-T and the control construct SP-eGFP-TurboID (eGFP-T) was confirmed, validating their suitability for PL experiments. Efficient TurboID-mediated biotinylation was achieved in the apoplast using co-infiltration of biotin, ATP and magnesium acetate. Streptavidin-HRP immunoblotting revealed distinct biotinylation profiles for INF1-T and eGFP-T. Furthermore, co-immunoprecipitation demonstrated specific biotinylation of REL by INF1-T, but not by eGFP-T, in wild-type, bak1 and sobir1/sobir1-like N. benthamiana. These findings demonstrate that TurboID-based PL is functional in the apoplast and provides a proof-of-concept for investigating elicitor-receptor interactions in this compartment.

Reactive oxygen species (ROS) are increasingly recognized as important oxidants in soil redox processes, with microbially produced ROS contributing broadly to elemental transformations. Given the active methane oxidation in paddy soils, methanotrophs have considerable potential to generate ROS yet their role in As(III) oxidation remains unclear. Here, we identify methanotroph-derived ROS as an important yet previously overlooked pathway for As(III) oxidation in paddy soils. Soil microcosm experiments and pure culture experiments revealed that methane oxidation enhanced ROS-mediated As(III) oxidation, with H2O2 identified as the dominant ROS. Inhibitor assays and functional gene analyses indicated that oxygen activation and respiratory electron transfer were closely associated with ROS generation during methane oxidation. These findings reveal methanotroph-derived ROS as a distinct route contributing to As(III) oxidation beyond dedicated arsenite oxidases. By revealing an overlooked connection between methane metabolism and arsenic transformation, this work expands our understanding of arsenic biogeochemistry and provides a conceptual basis for the coordinated mitigation of methane emissions and arsenic toxicity in agroecosystems.

While bacterial sensor histidine kinases (SHKs) are widespread as natural molecular biosensors, tools for high-throughput characterization of SHK signaling phenotypes are limited, hindering widespread implementation of bacterial-based sensing. Here, we developed a synthetic two-component signaling system that reports chimeric SHK signaling through a standardized fluorescence readout. With this synthetic system, we screened a library of chimeric DcuS/EnvZ SHKs to characterize sequence-function relationships within the DcuS sensory and transmembrane domains. We quantified the effects of 1173 mutations on signaling outputs in the presence of fumarate, a native DcuS ligand, as well as aspartate for which DcuS has minimal affinity for. We identified eleven positions across the DcuS domains which significantly alter aspartate responsiveness and additionally reported a role for cytoplasmic N-terminal residues in determining SHK signaling outputs. In future studies, this framework will expedite design of biosensors for novel ligands by enabling high-throughput screening of mutagenized libraries of natural SHKs.

The timely identification of drug-resistant pathogens and rapid differentiation of antibacterial mechanisms are critical in combating the escalating crisis of antimicrobial resistance. However, traditional broth culturing methods are inherently time-consuming and lack single-cell sensitivity. In this study, we reported a highly responsive biosensing platform based on single-entity impact electrochemistry (SIE) to rapidly differentiate bacterial activity at the single bacterium level. This method utilized p-benzoquinone (BQ) and BQ-K3Fe(CN)6 as redox mediators that were continuously reduced by the bacterial respiratory chain. Upon the collision of an individual bacterium with the ultramicroelectrode, these mediators were re-oxidized, generating transient current spikes that serve as a direct indicator of cellular viability. By tracking these distinct electrochemical signals, we successfully discriminated the responses of E. coli to three antibacterial agents. Specifically, the heavy metal stress induced by Pb2+ exposure resulted in a progressive decline in collision spike amplitude, with the total transferred charge decreasing from (7.51 ± 0.48) × 10–2 pC to (2.50 ± 0.14) × 10–2 pC per bacterium. The bacteriostatic effect of doxycycline yielded negligible immediate signal change due to its protein synthesis inhibition mechanism. In contrast, the membrane-disrupting effect of polymyxin B triggered a rapid cessation of collision signals alongside an increase in total current amplitude due to the leakage of intracellular redox contents. Based on these unique electrochemical signatures, we successfully identified polymyxin B-resistant E. coli mutants and achieved rapid validation of alternative antibiotics against this drug-resistant strain. This work expanded the analytical capabilities of SIE, offering a highly efficient platform for bacterial phenotypic screening and drug resistance monitoring.

Despite sharing nearly identical enzymes, strains of the same microbial species display wide variability of growth rates when growing on the same nutrients. Here, we show that such growth rate variability can emerge as a consequence of evolutionary adaptation to different patterns of environmental fluctuations. We develop a mathematical model combining cellular proteome allocation with eco-evolutionary dynamics in fluctuating environments. We find that different patterns of environmental fluctuations select for distinct growth profiles. Slow-growing strains repeatedly outcompete fast-growers when evolved in rapidly changing environments. In a range of environments with asymmetric nutrient availability, evolution reproducibly leads to the robust diversification of two coexisting strains. We develop a theoretical framework based on adaptive dynamics which quantitatively predicts these evolutionary outcomes and explains them in terms of the growth-lag tradeoff that all cells face. Finally, we show that the model can reproduce several patterns in growth-rate data from closely-related strains. Our results highlight that bacterial growth rates can rapidly adapt to their recent environmental history by proteome allocation alone, in ways that often oppose expectations based on nutrient quality.