 Your new post is loading...
|
Scooped by
mhryu@live.com
Today, 3:03 PM
|
Purple non-sulfur bacteria (PNSB) are metabolically versatile microorganisms that inhabit diverse environments taking advantage of a remarkably efficient photosynthetic machinery. In this work, we describe the creation of a workflow for experimental studies aiming to compare the light absorption properties of structural variants of the light-harvesting complex 2 (LH2) of PNSB. Through the creation of a library of LH2 variants using site-directed mutagenesis, we engineered proteins with different spectral properties to be expressed in a LH2-defficient mutant of the model PNSB Rhodobacter sphaeroides. We validated this approach by reproducing a previously described mutant with a blue-shift in the B850 absorption peak, in addition to identifying a novel mutant exhibiting a red-shift of this peak. We characterized the fluorescence lifetime of the purified LH2 spectral variants in vitro. Our results suggest that directed evolution approaches can be used to systematically explore the landscape of light absorption in LH2s variants.
|
Scooped by
mhryu@live.com
Today, 2:59 PM
|
PromA plasmids are a broad-host-range plasmid group and have been isolated from various environmental samples by exogenous plasmid capturing methods. However, their natural hosts have remained unidentified. Here, we developed a cultivation-independent single-cell-level screening approach combining water-in-oil droplets and digital PCR with a fluorescent probe, which enabled direct detection of PromAγ-related repA signals associated with environmental bacterial taxa. The positive droplets were collected by a dispenser with microfluidics and the potential natural hosts of the plasmid were identified based on their 16S rRNA gene sequences. Application to Lake Sanaru sediment samples collected on two dates identified diverse bacterial taxa associated with PromAγ-related repA signals, including previously unreported candidate hosts. Our results suggested that PromAγ plasmids were associated with a wide variety of bacterial taxa in natural microbial communities. The developed methods could be applied for the detection and collection of bacterial cells with a targeted gene, which would be valuable for microbial ecology studies.
|
Scooped by
mhryu@live.com
Today, 2:52 PM
|
A useful concept in microbial ecology is the “core microbiome.” Typically, core microbiomes are defined as the microbial taxa, genes, or functions shared by a threshold number of microbiome samples from a particular type of habitat (e.g., a particular type of host or a particular type of environment/ecosystem). In defining the core microbiome, the goal is to capture the portion of the microbial community that is conserved across samples from the focal habitat. Recently, there has been growing interest in developing methods to better characterize core microbiomes. As a result, numerous occurrence- and abundance-based measures have been defined. However, few have included phylogeny-aware metrics for analyzing core microbiomes. In this paper, we develop the concept of the “core community phylogeny”—a phylogeny where branches are selected based on their presence in multiple samples from a single type of habitat. We then use the core community phylogeny to define phylogenetic metrics describing the diversity of core microbiomes from a single type of habitat, the turnover between core microbiomes from two different types of habitats, and the shared diversity across core microbiomes from two or more different types of habitats. As compared to non-phylogenetic metrics, our phylogenetic metrics show greater consistency across taxonomic rank and/or phylogenetic level, as well as less sensitivity to strain variation across microbiome samples. Thus, our metrics address key challenges in the interpretation of core microbiomes. We provide a phylogenetic framework for characterizing and comparing core microbiomes. Importantly, the methods that we propose allow seamless integration of microbiome properties across taxonomic rank and/or phylogenetic level. Ultimately, this will provide both a more consistent picture of the core microbiome, as well as novel biological insight into the conserved components of microbial communities.
|
Scooped by
mhryu@live.com
Today, 2:45 PM
|
Streptomyces bacteria harbor conjugative plasmids and integrative conjugative elements that employ a unique mode of DNA transfer. It minimally requires a single FtsK-like motor protein, TraB, and a cis-acting sequence recognized by TraB, clt. Notably, clt-like sequences (clc) are also scattered over Streptomyces chromosomes, driving the co-transfer of their macrofragments along with the plasmid. This could be used to facilitate the exchange of genetic material between different streptomycete strains and even species, thus creating new opportunities for natural product discovery. Here we explored several possible avenues of use of conjugation in the context of the popular chassis strain Streptomyces albidoflavus J1074, such as plasmid and chromosomal marker mobilization, and ability to transfer across species. First, among several clc types being screened in silico, the one for TraB encoded by plasmid pIJ101 was the most abundant. Second, TraB-mediated transfer of replicative (pIJ101-based) plasmid occurred at a frequency of 10% of recipient spore titer. The efficiency of inheritance of the chromosomal marker, distant to the conjugative plasmid location within the genome, was about four orders of magnitude lower. The developed methods might be useful to produce novel J1074 genotypes. Finally, we show that the removal of native plasmid-encoded clt site and repressor gene korA led to further drop (down to 10− 6) in conjugation efficiency, albeit did not abrogate it completely. Hence, the prospects of chromosomal clc sites for gene transfer across species deserve more attention. We discuss the challenges associated with this genetic system and possible avenues for improvement, in the context of ongoing chassis-centered natural product discovery endeavors.
|
Scooped by
mhryu@live.com
Today, 2:22 PM
|
The cytosolic delivery of therapeutic macromolecules—including genome editors, transcription factors, and immunomodulators—remains a critical bottleneck in precision medicine, as most synthetic carriers rely on inefficient endocytic uptake. The bacterial type III secretion system (T3SS) can be engineered as a class of programmable living bioactive nanomaterials capable of direct, endocytosis-independent cytosolic injection. This review builds a unified engineering framework integrating three orthogonal modules: biosafety-optimized bacterial chassis, cargo–nanoconduit interfacial engineering to improve translocation competence, and synthetic sensing circuits enabling spatiotemporal cargo release, with the overall aim of preserving cargo functionality after delivery. To standardize cross-platform evaluation, the functional efficacy index (Φfunc) is introduced as a stage-resolved metric integrating secretion, conformational recovery, and activation efficiencies for rational living nanomaterial design. T3SS nanomaterials are benchmarked against lipid nanoparticles (LNPs), adeno-associated viruses (AAVs), and extracellular vesicles (EVs) across nine performance dimensions, suggesting advantages in cargo-dependent delivery of selected large proteins, genetic programmability, and the potential for repeated administration, subject to host immunity and chassis safety. Three application areas are outlined—cancer immunotherapy, genome/protein replacement therapy, and mucosal vaccination—and a preclinical-to-clinical development framework is proposed to address key barriers including immune clearance, heterogeneous tumor targeting, and biosafety. By reframing T3SS research under bioactive-material design principles, this review provides a unified evaluation framework to guide the rational development of living therapeutics toward clinical translation.
|
Scooped by
mhryu@live.com
Today, 1:38 PM
|
Base editing enables precise nucleotide substitutions without introducing DNA double-strand breaks, but identifying optimal editor-guide combinations remains challenging because PAM compatibility, editing-window positioning, bystander modifications, and off-target effects must be considered simultaneously. Here, we introduce BEguided, an integrated computational platform for variant-specific gRNA design and systematic comparison across 38 predefined base-editor profiles, with support for user-defined editors. BEguided incorporates single-nucleotide variants directly into target sequences, eliminating the need for customized reference genomes, while its companion tool, BEguidedOFF, provides base-editing-aware off-target prediction. To demonstrate the utility of this approach, we systematically assessed the accessibility of the 1,000 most frequent ABE-editable cancer mutations to adenine base editing. Under stringent guide-quality, bystander-editing, and predicted off-target criteria, only 7 mutations could be targeted using SpCas9-ABE8e. Expanding the analysis to additional SpCas9-compatible adenine base editors increased this number to 30, while inclusion of SaCas9, relaxed-PAM Cas9-NG/SpG variants, and PAM-less SpRY further expanded the number of potentially targetable mutations to 100, 156, and 238, respectively. These findings highlight the substantial impact of editor selection on mutation accessibility and establish BEguided as a versatile framework for identifying and prioritizing base-editing strategies for disease-associated variants. BEguided is available at https://beguided.buchholzlab.org
|
Scooped by
mhryu@live.com
Today, 1:15 PM
|
Thermoset composites have become indispensable high-performance materials across key industrial sectors. However, existing recycling methods have yet to deliver sustainable end-of-life solutions for these materials. Here, we report the engineering of a highly active and stable biocatalyst capable of depolymerizing polyester-based thermoset composites. First, a second-generation polyester resin was designed to enhance susceptibility to enzymatic attack. Cutinase from Fusarium solani was selected and subsequently engineered through a combination of directed evolution and computational design using two soluble surrogate motifs representative of the redesigned resin structure. The resulting variant, which incorporated 19 mutations, exhibited a 22-fold increase in activity and an 11°C improvement in melting temperature while achieving recombinant production titers at the gram-per-liter scale in yeast cultivated in fed-batch bioreactor. This engineered enzyme depolymerized the polyester resin with a total release of 6 mM (∼1.08 grams per liter) monomethyl phthalate. Together, to our knowledge, these findings constitute the first documented example of enzymatic degradation of a complex polyester resin from thermoset composite materials.
|
Scooped by
mhryu@live.com
November 20, 5:13 PM
|
E. coli employs stress responses to maintain homeostasis and mitigate harmful impacts on cellular and genomic integrity. Two key systems managing tolerance to unfavorable conditions include the DNA damage-inducible SOS response and the oxidative stress-inducible OxyR response. Recently, we demonstrated that simultaneous activation of these two systems resulted in the suppression of the SOS-associated hypermutability, suggesting that these systems when co-induced may yield new regulatory features. To further investigate these systems individually and jointly we used whole-transcriptome sequencing and proteomics to assess the underlying gene and protein networks. We used E. coli K-12 strain MG1655 and its derivatives carrying the recA730 and oxyR2c alleles that yield constitutive expression of the SOS and OxyR systems, respectively. Our findings indicate the genetic scope of these systems to be significantly broader than previously reported, revealing scores of differentially expressed genes not formerly classified as part of the systems (143 in the recA730 and 248 in the oxyR2c strain). A key finding was that the constitutive co-expression of the two systems led to a broad suppression of genes of both systems. Importantly, a novel group of 137 genes not belonging to either individual system proved differentially expressed under co-activation conditions, which was further corroborated by proteomic data. Overall, the data suggest that SOS-induced cells may be preoccupied with aerobic energy producing activities, while OxyR-induced cells are focused on energy-conserving reactions, biofilm formation and reducing motility. Under co-activating conditions a compromise might be reached that represents a further adaptive mechanism to support survival under multi-stress conditions.
|
Scooped by
mhryu@live.com
November 20, 5:04 PM
|
Ruminant gut microbial communities influence host health, productivity, and methane emissions, yet viral communities across life stages remain uncharacterized. Whilst bacteriophages shape microbial communities in many ecosystems, their role across dairy cattle life stages is unknown. We characterized DNA viral community composition, diversity, and ecology in cows belonging to four life stages: calf, heifer, dry adult, and lactating adult. Using hybrid Illumina and Oxford Nanopore sequencing, we assembled 30,321 viral operational taxonomic units from fecal samples, including 1338 complete genomes (92% representing novel genera). Viral communities showed distinct life-stage stratification. Calves harbored low-diversity communities (Shannon index: 2.49 vs >6.58 in adults). They also showed a higher proportion of temperate phages (49% vs <24% in adults), though this difference was not statistically significant given the small calf sample (n = 3). Adult viromes showed 50-fold higher community evenness than calves. Viral and bacterial diversity were inversely correlated (Spearman’s rho = −0.69), with viral diversity peaking during the drying-off period when bacterial diversity was lowest. Dry cows showed the highest viral loads (3.61% of microbial DNA vs 1.52% in lactating cows) and elevated virus-host ratios, coinciding with increased positive selection on viral genes. Twenty-six vOTUs targeting the methanogen Methanobrevibacter were detected in adults but not calves, paralleling a 757-fold higher relative abundance of Methanobrevibacter in adults. Additionally, we identified viruses carrying putative auxiliary metabolic genes involved in methane metabolism pathways. The dairy cattle gut virome is extensive, almost entirely novel, and clearly stratified by life stage—characterised here across calves, heifers, dry adults, and lactating adults for the first time. Viral and bacterial diversity showed a significant inverse relationship that was most pronounced around drying-off, and viruses targeting the methanogen Methanobrevibacter were present in adults but absent from calves. Although cross-sectional, the data establish life stage as a key variable in the ruminant gut virome and identify methanogen-infecting phages as a candidate for reducing enteric methane emissions.
|
Scooped by
mhryu@live.com
November 20, 4:54 PM
|
Temperate and lytic phages can redirect host-associated bacterial phenotypes through distinct mechanisms. Lytic phages may attenuate virulence when resistance carries pleiotropic costs, whereas prophage-encoded regulation can directly enhance bacterial fitness. We propose that these outcomes represent complementary routes of phage-mediated phenotypic filtering, determined by how phage susceptibility, resistance pathways, or phage-encoded functions intersect with bacterial traits governing colonization, immunity, and disease.
|
Scooped by
mhryu@live.com
November 20, 3:09 PM
|
Understanding the ecological and epidemiological dynamics of microbial populations requires resolution of genetic diversity beyond the species level, yet conventional culture-based surveys and standard sequencing of metagenomic samples rarely provide sufficient on-target depth for variant-level analysis. Here, we present a generalizable molecular and computational framework for variant-resolved analysis of specific microbial taxa within complex microbial communities. The workflow integrates a customized capture-probe design, an optimized target-enriched (TE) shotgun sequencing strategy, and a dedicated computational approach for variant-level inference from metagenomic data. The computational approach constructs a genomic sequence variant (GSV) database for the targeted taxa using reference genomes clustered by average nucleotide identity, which is used in the bioinformatic pipeline (VARIANT++) for pseudoalignment and probabilistic variant inference. We demonstrate the performance of this framework using Mannheimia haemolytica (Mh) as a model pathogen target. Targeted sequencing using a custom probe set designed from 70 reference genomes, which recovers >97% of Mh genomes not used in its design, achieved >380-fold enrichment of Mh reads relative to shotgun sequencing of the same samples, enabling variant-level resolution even when the target comprised a small fraction of total community DNA. Application of the workflow to bovine respiratory samples revealed extensive within-animal genomic diversity of the target pathogen, with most samples containing multiple co-occurring GSVs, indicating that mixed-variant populations are a common feature of Mh ecology that cannot be resolved using conventional sequencing approaches. Collectively, these results demonstrate that TE metagenomic sequencing coupled with variant-aware bioinformatics provide a scalable and adaptable approach for resolving fine-scale population structures and transmission dynamics of specific microbial taxa within complex microbial communities.
|
Scooped by
mhryu@live.com
November 20, 2:41 PM
|
Phage lytic proteins are promising antibacterial and biotechnological candidates, yet uncertain annotations, modular architectures, and class imbalance complicate sequence-based classification. We developed PhageLysClass, an evidence-aware framework for recognizing lytic proteins and distinguishing Endolysins, Depolymerases, and virion-associated lytic enzymes. From 442,329 unique phage sequences, including 11,867 evidence-supported positives, we defined distinct candidate non-target, unlabeled, and challenge populations. Seven supervised and two positive-unlabeled tasks were evaluated across factorial redundancy and train-test homology conditions, yielding 2,240 and 768 experiments, respectively. Redundancy reduction influenced generalization more strongly than homology separation, with non-additive interactions between them. ESM2-T30 consistently outperformed alternative representations and supported a hierarchical classifier achieving approximately 0.89 MCC under strict R30/S30 conditions in the selected, pre-refitting benchmark. Applied without retraining to 77,897 sequences across three collections, PhageLysClass combined predictions with physicochemical and geometric contextualization, highlighting Depolymerase-like candidates among structurally ambiguous proteins while distinguishing reference-supported assignments from extrapolation.
|
Scooped by
mhryu@live.com
November 20, 1:32 AM
|
Riboflavin is an essential vitamin widely used in pharmaceutical and feed applications, and its industrial production relies primarily on microbial fermentation. However, riboflavin biosynthesis in E. coli is constrained by precursor availability and broader physiological regulation. In this study, a small regulatory RNA (sRNA)-mediated attenuation strategy was employed to systematically screen 91 genes across five functional modules. Twenty-three targets increased riboflavin production by more than 20%, and five key targets were further validated through genome-level modification. Combinatorial attenuation of rpe and yrfG, optimization of the nrfA-targeting sRNA expression cassette, and chromosomal reinforcement of the riboflavin operon were subsequently integrated to construct strain SN49M. Following fermentation medium optimization, SN49M produced 3153.3 mg/L riboflavin at 46 h in shake-flask cultivation with a yield of 193.4 mg/g glucose. In fed-batch fermentation, SN49M reached 20.5 g/L riboflavin within 72 h, with an overall yield of 178.9 mg/g glucose, representing the highest reported glucose-based riboflavin yield achieved by microbial fermentation. This work establishes a systematic and tunable framework for discovering and integrating metabolic and physiological perturbations, providing an effective strategy for improving carbon efficiency in microbial riboflavin production.
|
|
Scooped by
mhryu@live.com
Today, 3:02 PM
|
Engineered living materials (ELM) combine genetically programmable cells with material scaffolds to deliver adaptive functions such as sensing, secretion, and self-repair. However, their deployment beyond the laboratory remains constrained by a gap between proof-of-concept performance and robust operation in complex real-world environments. Here, we present a chassis-centered review of deployable ELMs, focusing on how engineered cells can be equipped to survive environmental stress, sustain functional performance during prolonged operation, and remain safely controlled in open or host-associated settings. This perspective defines key design priorities for building safer, more durable, and application-specific living materials. By integrating these principles, we propose a robustness-by-design framework for selecting and engineering cellular chassis as ELMs transition toward reliable biomedical, environmental, and industrial deployment.
|
Scooped by
mhryu@live.com
Today, 2:57 PM
|
Microcystins are among the most prevalent cyanobacterial toxins threatening freshwater ecosystems and human health, yet the molecular basis of their bacterial detoxification remains incompletely understood. Here, we present an integrated structural model of the MlrABCD pathway from Sphingopyxis sp. C-1, providing mechanistic insight into the sequential transformation of microcystins across distinct bacterial cellular compartments. Comparative genomics, structural modelling, ligand-binding predictions, and molecular dynamics simulations reveal a spatially coordinated detoxification process in which membrane-associated MlrA initiates toxin linearization, periplasmic MlrB hydrolyses linear intermediates, MlrD transports degradation products across the cytoplasmic membrane, and cytoplasmic MlrC completes peptide degradation. Analysis of environmentally relevant microcystin congeners showed that substrate recognition is governed by the interplay between toxin chemistry and enzyme architecture. MlrA preferentially recognizes MC-LR, whereas MlrB displays enhanced affinity for linearized MC-RR, indicating that ring opening reshapes substrate selectivity during pathway progression. Substrate binding also stabilizes MlrD, supporting a ligand-induced alternating-access transport mechanism. Collectively, these findings indicate that the Sphingopyxis sp. C-1 biodegradation machinery has evolved under environmental selective pressures into a compartmentalised and highly coordinated detoxification system optimized for the predominant environmental congeners, MC-LR and MC-RR, providing a molecular framework for the rational development of enhanced cyanotoxin bioremediation strategies.
|
Scooped by
mhryu@live.com
Today, 2:46 PM
|
The last decade has brought renewed interest to microbial influences on host biology, yet the roles of microbes in host reproductive biology are underappreciated and often defined by a profound maternal bias. This perspective synthesizes emerging evidence across the tree of life to highlight the paternal microbiome as an overlooked driver of fitness, reproductive output, and intergenerational health, impacting epigenetic sculpting of the sperm and embryo, sexual transmission of microbes, and male gamete functions. These contributions indicate that mating can involve interactions between holobionts rather than simply the union of the maternal and paternal genomes. Thus, uniting male reproduction with the growing generalizations of microbial symbiosis guides several conceptual, theoretical, and empirical avenues for research in the life sciences.
|
Scooped by
mhryu@live.com
Today, 2:41 PM
|
Prime editing (PE) supports versatile genome modifications, yet high-efficiency PE3 and PE5 systems nick opposite DNA strands, which can occasionally lead to double-strand breaks (DSBs), thereby increasing editing efficiency but also raising the risk of DNA damage. We hypothesized that an additional nick on the edited strand facilitates 5′-flap processing, thereby improving PE efficiency. Using this single-strand dual-nick PE (sdPE) strategy, editing performance was greatly enhanced across tested loci: sdPE2, sdPE3, sdPE4max, and sdPE5max outperformed their parental PE systems at 17/20, 14/20, 14/20, and 9/20 loci, respectively. Two-guide RNA (gRNA) (one PE gRNA [pegRNA] plus one additional single gRNA [sgRNA]) sdPE2/sdPE4max achieve comparable or slightly lower efficiency than conventional two-gRNA PE3/PE5max but with lower indel frequencies. Moreover, three-gRNA (one pegRNA plus two additional sgRNAs) sdPE3b/sdPE5bmax further improves the performance of PE3b and PE5bmax. Thus, sdPE is compatible with most existing PE systems and promotes favorable editing outcomes by introducing a tailored sgRNA targeting the edited strand.
|
Scooped by
mhryu@live.com
Today, 2:07 PM
|
Bacteria use sophisticated mechanisms to defend themselves against antibiotics and bacteriophages, yet how they combat toxins deployed during interbacterial competition remains poorly understood. Bacteria use type VI secretion, bacteriocins, and contact dependent inhibition systems to deliver antibacterial toxins that inhibit growth by antagonizing essential processes. Here, we discover that Pseudomonas aeruginosa uses homologs of the eukaryotic major vault protein and a tRNA-like noncoding RNA to defend against RNA-damaging ribotoxins. Using an engineered pyocin delivery system, we showed that P. aeruginosa sensed ribotoxin damage and activated transcription of an RNA repair operon, which included band7, a homolog of the major vault protein. Genetic analyses demonstrated that band7-expressing bacteria were significantly more protected against growth inhibition by a ribotoxin that cleaved the anticodon loop of specific tRNA. Band7 was necessary but not sufficient, and defense also required a non-coding band7 RNA (bRNA). bRNA was cleaved by the ribotoxin, suggesting a mechanism for specificity and defense. Bioinformatic analyses revealed that band7 and candidate bRNA are widely distributed within RNA repair systems across diverse bacteria. Our data suggest that bacteria use structures related to the eukaryotic vault complex to respond to RNA damage thereby providing a hypothesis for the function of the enigmatic human vault.
|
Scooped by
mhryu@live.com
Today, 1:29 PM
|
Reaction-diffusion (RD) has long been proposed as a mechanism underlying biological pattern formation. However, quantitative frameworks linking circuit design to spatial organization remain limited. Here, we engineered a synthetic gene circuit, MINPAC, that couples gene expression regulation (reaction) with bacterial colony expansion and quorum sensing (diffusion) to guide bacterial cells to self-organize into periodic volumetric patterns across multiple scales, from single cells to three-dimensional colonies. The patterns exhibit robustness against environmental variations and size-dependent complexity. Mathematical modeling supports a clock-and-wavefront mechanism, where oscillatory gene expression interacts with outward expanding diffusions to generate these periodic spatial structures. Topological analysis reveals that specific network architecture is essential for pattern formation. Guided by model predictions, we further show that pattern complexity can be systematically tuned by gene dosage, with lower expression levels favoring multiple rings through faster dynamics. This work establishes a framework uniting circuit engineering and mathematical analysis for programming multicellular self-organization through RD mechanisms, with potential applications in regenerative medicine and biomaterials.
|
Scooped by
mhryu@live.com
Today, 12:59 PM
|
Predicting the dynamics of ecological systems is a central challenge in biology. One factor that may limit predictability is ecological chaos. Theory has long suggested that even simple interactions can produce chaos, yet empirical demonstrations in living systems remain scarce. In a two-strain system of E. coli, increasing the duration between resource replenishment reveals transitions from competitive exclusion to chaos, driven by asymmetric ecological interactions and periodic environmental forcing. Guided by these observations, we developed a dynamical systems framework grounded in empirical data that recapitulate the observed dynamics and predicts intermediate ecological states and transitions. Experimental validation of these predictions by implementing the corresponding conditions successfully reproduces the full range of predicted dynamical regimes. Because feast-famine cycles and asymmetric interactions are widespread in nature, our findings provide a model for uncovering the ecological and evolutionary significance of chaos, opening avenues to use resource fluctuation and interaction strength to quantify, monitor, and manipulate complex ecosystem behavior previously interpreted as stochasticity.
|
Scooped by
mhryu@live.com
November 20, 5:10 PM
|
Emerging tools for spatial analysis, systems biology, and synthetic biology offer opportunities to observe, dissect, and perturb cancer mechanisms, to design strategies for the prevention of oncogenesis, and to identify therapies that lead to durable tumor control. Here, we summarize perspectives from the virtual Innovation Lab on “Synthetic and Systems Approaches to Interrogate Spatiotemporal Processes in Cancer” (S3) hosted by the National Institutes of Health (NIH), which assembled interdisciplinary experts to discuss cancer spatial organization and dynamic cellular interactions and how they can be addressed by leveraging novel synthetic and systems biology approaches. These discussions focused especially on cancer initiation, which results from a failure of systems control. To catalyze innovation across the research community, we summarize key insights from these events into a vision for integrated synthetic, systems, and spatial approaches in cancer research.
|
Scooped by
mhryu@live.com
November 20, 4:58 PM
|
Virus-host association prediction is fundamental to virus-related disease control and phage therapy, yet many existing tools require candidate-host genome sequences, which may be unavailable for some eukaryotic hosts or impractical to search at the scale of large virome studies. We present VHSeek, a supervised virus-host association predictor that integrates DNA- and protein-derived viral representations. VHSeek combines nucleotide representations from AIDO.DNA and protein representations from ESM2 to learn virus-host association patterns, and then transfers predictions through known host taxonomies from lower to higher ranks. During inference, VHSeek uses viral sequence data and known host taxonomies, without host sequence data. Compared with iPHoP, the state-of-the-art baseline evaluated in this study, VHSeek improves genus-level maximum F1 score (Fmax), Top-1 accuracy, and area under the precision–recall curve (AUPR) by 12%, 19%, and 27%, respectively. Confidence thresholding enables abstention and reduces false discoveries. On the processed IMG/VR v4.1 metagenomic collection, VHSeek expands effective host coverage from 36.5% to 64.5% using a validation-calibrated confidence threshold. VHSeek prioritizes candidate viruses and reservoir hosts with potential human health impacts, including multi-host hypotheses in which one virus may be associated with humans and additional hosts (e.g., Saboya virus with Homo sapiens and Culicidae) not currently captured in the dataset. The authors present VHSeek, a tool that uses DNA and protein sequence representations to predict viral hosts and prioritize candidate virus-host associations across large viral collections.
|
Scooped by
mhryu@live.com
November 20, 4:53 PM
|
As genome resources continue to expand, estimating the accuracy and completeness of genome annotations (e.g., gene and protein models) becomes essential to ensure quality of downstream analyses and applications. It becomes critical in comparative genomic analysis to select genomes with comparable qualities of assembly and annotation. Without careful scrutiny, inconsistencies in annotation quality can undermine downstream studies. We have developed Accuracy-Based Annotation Quality Score (ABAQS), a novel method to comprehensively assess genome annotation quality. ABAQS evaluates multiple factors, including genome completeness, gene model validity, and protein profile accuracy, providing a more complete picture of annotation quality than metrics like BUSCO and PSAURON. We applied ABAQS to 1700+ eukaryotic fungal and algal genomes and showed its robustness and effectiveness in evaluating genome annotation quality, making it a valuable tool for researchers working with genomic data.
|
Scooped by
mhryu@live.com
November 20, 3:08 PM
|
Poly(ethylene terephthalate) (PET) is one of the most widely used synthetic plastics, yet its environmental accumulation continues to outpace effective recycling. Although PET-degrading enzymes and microorganisms have been reported, discovery of new PET-degrading biological resources remains slow because conventional screening depends on labor-intensive cultivation and chromatographic analysis. A sensitive whole-cell biosensor that reports terephthalic acid (TPA), a key PET hydrolysis product, could substantially accelerate the identification of PET-degrading microbes from environmental samples. Here we engineered a TPA-responsive whole-cell sensor by evolving the Acinetobacter baylyi PobR transcription factor (TF), which naturally responds to 4-hydroxybenzoate but not TPA. A biosensor composed of PobR, its cognate promoter region, GFP, and the evolved A. baylyi MucK transporter did not respond to TPA in its wild-type form. We therefore performed combinatorial library construction and screening of PobR using site-saturation mutagenesis at ten ligand-recognition positions and fluorescence-activated cell sorting. The final variant, designated PobR_v10, contained the substitutions S118T, ΔL141, V143Y, M148A, T159I, S160T, S212A, H216F, L220M, and M241L, and detected TPA over a dynamic range of 50–500 μM with a detection limit of 50 μM. To convert this biosensor into a colony-screening platform, we coupled the TPA-responsive circuit to the Bacillus subtilis racE gene in a D-glutamate auxotrophic E. coli host, enabling inducer-dependent growth and fluorescence through cell-to-cell communication on solid medium. The resulting sensor selectively visualized PET hydrolysis around colonies expressing leaf-branch compost cutinase and enabled rapid screening of environmental samples on bis(2-hydroxyethyl) terephthalate (BHET)-containing agar. Five candidate isolates were recovered and validated by HPLC-based assays. Among them, the isolate identified as Rhodococcus qingshengii showed the strongest activity toward PET, BHET, and MHET and exhibited robust growth with PET as the sole added carbon source. Unlike previously reported TphR-based systems, the engineered PobR platform was directly integrated into a growth-coupled spatial screening architecture compatible with colony-level PET hydrolysis detection. This study establishes an engineered PobR-based TPA biosensor and demonstrates its application as a rapid cell-to-cell communication platform for discovering PET-degrading microbes on solid media. The system expands the repertoire of TPA-responsive TF sensors and provides a practical engineering tool for environmental bioprospecting, enzyme discovery, and future PET upcycling workflows.
|
Scooped by
mhryu@live.com
November 20, 2:28 PM
|
Synthetic biology allows for the synthesis of genomes from scratch. Although technically achievable, our abilities in designing genomes that result in live organisms are in their infancy. One important consideration is synonymous codon usage bias, which is present in most organisms and shaped by selection for efficient translation. Many studies have edited the synonymous codon usage patterns of viruses with the goal of viral attenuation, but these studies almost always build a single construct per design, so we know little about how repeatable the effects of recoding are. We designed and built 111 synonymously recoded ΦX174 genomes, applying four recoding strategies at up to five levels of attenuation to each of three genes, with three independently designed replicates of most combinations. Which gene we recoded mattered considerably more than how we recoded it. Recoding genes for the major capsid protein F, the major spike protein G and the DNA pilot protein H in full, with the least-used synonym codons from E. coli cost 12.0, 10.8 and 5.9 doublings per hour, respectively. Recoding strategy per se had no detectable effect once the number of codons changed was accounted for. Predicted mRNA folding stability also had no measurable effect on fitness when tested in gene F, whose wildtype transcript is highly stable and could easily be destabilized with synonymous codon rearrangements. While the number of codons changed was correlated with the degree of attenuation, divergence between independently designed replicates increased as more codons were edited. Replicates of designs with small fitness effects varied little, while replicates of strongly attenuating designs diverged by several doublings per hour. The fitness outcome of large-scale genome recoding is therefore substantially unpredictable from the design variables alone: even among independently designed genomes receiving the same treatment, fitness can diverge substantially.
|