N Nature Biotechnology · Dec 05, 2025 In vivo gene editing of human hematopoietic stem and progenitor cells using envelope-engineered virus-like particles Engineered virus-like particles (VLPs) are a promising technology for in vivo gene editing of human hematopoietic stem and progenitor cells (HSPCs). Here we design and test two different VLP envelopes for human HSPC editing in vitro and in vivo. The first is an optimized version of the baboon envelope BaEVTR, which efficiently transduces human HSPCs in vitro. We show that the optimized BaEVTR VLP enables in vivo editing of β2 microglobulin in long-term human HSPCs (31% at 8 weeks after dosing) and editing of two hemoglobinopathy-relevant loci,BCL11AandHBG1/2(26% and 7.5%, respectively, at 5 days after dosing), inducing fetal hemoglobin. Our second VLP design uses a CD133-targeted envelope designed to reduce the transduction of mature blood cells and achieves higher in vivo specificity for HSPCs compared to the optimized BaEVTR VLP. As avoiding delivery in filter organs such as the liver would enhance efficiency and safety, we also demonstrate that both VLPs avoid human hepatocytes in a humanized liver model. Gene therapy Genetic vectors Haematopoietic stem cells biology mouse experiments
N Nature Biotechnology · Dec 03, 2025 Mapping single-cell diploid chromatin fiber architectures using DAF-seq Gene regulation is orchestrated by the co-binding of proteins along chromosome-length chromatin fibers within single cells, yet the heterogeneity of this occupancy between haplotypes and cells remains poorly resolved in diploid organisms. Here we present Deaminase-Assisted single-molecule chromatin Fiber sequencing (DAF-seq), which enables single-molecule footprinting at near-nucleotide resolution while synchronously profiling single-molecule chromatin states and DNA sequence. DAF-seq illuminates cooperative protein occupancy at individual regulatory elements and resolves the functional impact of somatic variants and rare chromatin epialleles. Single-cell DAF-seq (scDAF-seq) generates chromosome-length protein co-occupancy maps across 99% of each individual cell’s mappable genome. scDAF-seq uncovers extensive chromatin plasticity both within and between single diploid cells, with chromatin actuation diverging by 61% between haplotypes within a cell, and 63% between cells. Moreover, we find that regulatory elements are preferentially co-actuated along the same fiber in a distance-dependent manner that mirrors cohesin-mediated loops. Overall, DAF-seq enables the characterization of protein occupancy across entire chromosomes with single-nucleotide, single-molecule, single-haplotype and single-cell precision. Epigenomics Genomics Sequencing biology
N Nature Biotechnology · Dec 03, 2025 Standardized metrics for assessment and reproducibility of imaging-based spatial transcriptomics datasets Spatial transcriptomics lacks standardized metrics for evaluating imaging-based in situ hybridization technologies across sites. In this study, we generated the Spatial Touchstone (ST) dataset from six tissue types across several global sites with centralized sectioning, analyzed on both Xenium and CosMx platforms. These platforms were selected for their widespread use and distinct chemistries. We assessed reproducibility, sensitivity, dynamic ranges, signal-to-noise ratio, false discovery rates, cell type annotation and congruence with single-cell profiling. This study offers ST standardized operating procedures (STSOPs) and an open-source software, SpatialQM, enabling evaluation of samples across all technical metrics and direct imputation of cell annotations. The generated imaging-based spatial transcriptomics data repository comprises 254 spatial profiles, incorporating both public and newly generated ST datasets in a web-based application, which enables analysis and comparison of user data against an extensive collection of imaging-based datasets. Finally, we establish best practices and metrics to evaluate and integrate imaging-based multi-omics data from single cells into spatial transcriptomics to spatial proteomics. Bioinformatics Quality control Standards Transcriptomics
N Nature Biotechnology · Nov 26, 2025 All-optical visualization of specific molecules in the ultrastructural context of brain tissue Understanding the molecular anatomy and neural connectivity of the brain requires imaging technologies that can map the three-dimensional nanoscale distribution of specific proteins in the context of brain ultrastructure. Light and electron microscopy visualize either specific labels or anatomical ultrastructure but combining molecular specificity with anatomical context is challenging. Here we present pan-expansion microscopy of tissue (pan-ExM-t), an all-optical imaging method that combines ~16–24-fold linear expansion with fluorescent pan-stainings of proteins and lipids (providing electron microscopy-like ultrastructural context) and immunolabeling (for molecular imaging). We demonstrate the versatility of this approach by imaging synaptic and cell-specific antibodies in the ultrastructural three-dimensional context of presynaptic and postsynaptic densities, neuropil nanoarchitecture and cellular organelles in dissociated neuron cultures, and mouse brain tissue sections. Furthermore, we demonstrate tracing of neuronal circuitry from pan-ExM-t image volumes, suggesting that any laboratory with access to a confocal microscope can now localize specific molecules within nanoscale cellular and circuit contexts. Cellular neuroscience Super-resolution microscopy biology mouse experiments
N Nature Biotechnology · Nov 26, 2025 Antimicrobial peptide delivery to lung as peptibody mRNA in anti-inflammatory lipids treats multidrug-resistant bacterial pneumonia The efficacy of antimicrobial peptides (AMPs) is limited by challenges of delivery and potency. We enhance AMP performance in the lung by converting AMPs to a peptibody format that fuses AMPs with fragment crystallizable domains to activate innate immunity and cathelin domains for infection-responsive activation, with their mRNA constructs delivered by anti-inflammatory lipid nanoparticles. The highest-scoring design outperforms antibiotic therapy approved by the US Food and Drug Administration in multidrug-resistant pneumonia models, eradicating representative MDR bacteria while mitigating inflammation. Drug delivery Nanomedicine biology
N Nature Biotechnology · Nov 25, 2025 Multiplexed profiling of transcriptional regulators in plant cells Transcriptional regulators play key roles in plant growth, development and environmental responses; however, understanding how their regulatory activity is encoded at the protein level has been hindered by a lack of multiplexed large-scale methods to characterize protein libraries in planta. Here we present enrichment of nucleartrans-elements reporter assay in plants with sequencing (ENTRAP-seq), a high-throughput method that introduces protein-coding libraries into plant cells to drive a nuclear magnetic sorting-based reporter, enabling multiplexed measurement of regulatory activity from thousands of protein variants. Using ENTRAP-seq and machine learning, we screen 1,495 plant viruses and identify hundreds of putative transcriptional regulatory domains found in structural proteins and enzymes not associated with gene regulation. In addition, we combine ENTRAP-seq with machine-guided design to engineer the activity of a plant transcription factor in a semirational fashion. Our findings demonstrate how scalable protein function assays deployed in planta will enable the characterization of natural and synthetic coding diversity in plants. Assay systems Functional genomics Molecular engineering in plants Plant molecular biology Transcriptional regulatory elements biology
N Nature Biotechnology · Nov 21, 2025 Programmable initiation of mRNA translation bytrans-RNA Several approaches exist to silence genes, but few tools are available to activate individual mRNAs for translation inside cells. Guiding ribosomes to specific start codons without altering the original sequence remains a formidable task. Here we design cappedtrans-RNAs capable of directing ribosomes to specific initiation sites on individual mRNAs when thetrans-cap is positioned near the target start codon. Structural and biochemical data suggest that the cappedtrans-RNA facilitates ribosome loading and scanning on the target mRNA through a synergistic mechanism involving alternative cap recognition. Thetrans-RNA also acts independently of the cap on the target mRNA, enabling translation of circular RNAs lacking internal ribosome entry sites. We applytrans-RNAs in vivo to achieve programmable alternative translation of endogenous genes in mouse liver. Finally, we provide the evidence for the existence of natural transcripts that, similarly to exogenoustrans-RNAs, activate translation of endogenous mRNAs. Ribosome RNA biology mouse experiments
N Nature Biotechnology · Nov 17, 2025 Extensive restoration of forelimb function in primates with spinal cord injury by neural stem cell transplantation Research on cell therapy for spinal cord injury has yet to achieve sufficient functional recovery. Previous studies in the field grafted oligodendrocyte progenitors, nonspinal neural stem cells or primary spinal neural progenitors. Here we sought to improve functional outcomes by grafting clinically compatible spinal cord neural stem cells derived from human embryonic stem cells (H9-scNSCs). H9-scNSCs significantly improved functional outcomes on a skilled hand task 9.2-fold (P= 2.5 × 10−27) in hemisected subjects compared with lesioned controls, achieving a fine object retrieval success of 53.4 ± 19.2%, and 2.9-fold (P= 6.3 × 10−8) superior to controls in hemicontused subjects. Recovery correlated with rehabilitation effort. Grafts extended up to hundreds of thousands of new axons into host circuits up to 39 mm below the injury, forming synapses with host circuitry. Lesion fill was substantially higher and differentiated cell-fate distributions were much closer to that of the normal spinal cord than in previous studies using primary spinal cord cells, likely enabling the observed superior functional outcomes. Embryonic stem cells Neural stem cells Spinal cord injury
N Nature Biotechnology · Nov 17, 2025 Fecal exfoliome sequencing captures immune dynamics of the healthy and inflamed gut Metagenomic sequencing and metabolomics of fecal matter have revealed the impact of the gut microbiome on health and disease. In addition to microbiota, feces also contain shed or exfoliated host epithelial, secretory and immune cells, but RNA profiling of these cells is challenging owing to degradation and cross-contamination. Here we introduce exfoliome sequencing (Foli-seq) to profile fecal exfoliated eukaryotic messenger RNAs (feRNAs) originating from the upper and lower gastrointestinal regions and show that this ‘fecal exfoliome’ harbors stable RNAs that reflect intestinal and immune function. By selectively amplifying targeted transcripts, Foli-seq demonstrates robust, accurate, sensitive and quantitative measurement of feRNAs. In murine colitis models, feRNA reveals temporal processes of epithelial damage, immune response and intestinal recovery specific to different types of gut inflammation. Simultaneous exfoliome and microbiome profiling uncovers a dense host–microbe interaction network. Moreover, we demonstrate stratification of patients with inflammatory bowel disease into subgroups that correlate with disease severity. Fecal Foli-seq is a noninvasive strategy to longitudinally study the gut and profile its health. Gene expression analysis Gene expression profiling Inflammation
N Nature Biotechnology · Nov 13, 2025 Isotropic, aberration-corrected light sheet microscopy for rapid high-resolution imaging of cleared tissue Light-sheet microscopy is ideal for imaging large and cleared tissues, but achieving a high isotropic resolution for a centimeter-sized sample is limited by slow and often aberrated, axially scanned light sheets. Here, we introduce a compact, high-speed light-sheet fluorescence microscope achieving 850 nm isotropic resolution across cleared samples up to 1 cm³ and refractive indices ranging from 1.33 to 1.56. Using off-the-shelf optics, we combine an air objective and a meniscus lens with an axially swept light sheet to achieve diffraction-limited resolution and aberration correction. The effective field of view is increased by twofold by correcting the field curvature of the light sheet using a concave mirror in the remote focusing unit. Adapting the light sheet’s motion with a closed-loop feedback enhances the imaging speed by tenfold, reaching 100 frames per second while maintaining resolution and field of view. We benchmark the system performance across scales, from subcellular structure up to centimeter scale, using various clearing methods. 3-D reconstruction Fluorescence imaging Light-sheet microscopy Neuroscience
N Nature Biotechnology · Nov 11, 2025 A parts list of promoters and gRNA scaffolds for mammalian genome engineering and molecular recording A standardized ‘parts list’ of sequences for genetic engineering of microbes has been indispensable to progress in synthetic biology, but few analogous parts exist for mammalian systems. Here we design libraries of extant, ancestral, mutagenized or miniaturized variants of polymerase III promoters and guide RNA (gRNA) scaffolds and quantify their abilities to mediate precise edits to the mammalian genome through multiplex prime editing. We identify thousands of parts for reproducible editing in human and mouse cell lines, including hundreds with greater activity than commonly used sequences. Saturation mutagenesis screens identify tolerated sequence variants that further enhance sequence diversity. In an application to molecular recording, we design a ‘ten key’ array that, in mammalian cells, achieves balanced activity of pegRNAs as predicted by the activity of the component parts. The data reported here will aid the design of synthetic loci encoding arrays of gRNAs exhibiting predictable, differentiated levels of activity for applications in multiplexed perturbation, biological recorders and complex genetic circuits. Functional genomics Genomics
N Nature Biotechnology · Nov 11, 2025 Multimodal learning enables chat-based exploration of single-cell data Single-cell sequencing characterizes biological samples at unprecedented scale and detail, but data interpretation remains challenging. Here, we present CellWhisperer, an artificial intelligence (AI) model and software tool for chat-based interrogation of gene expression. We establish a multimodal embedding of transcriptomes and their textual annotations, using contrastive learning on 1 million RNA sequencing profiles with AI-curated descriptions. This embedding informs a large language model that answers user-provided questions about cells and genes in natural-language chats. We benchmark CellWhisperer’s performance for zero-shot prediction of cell types and other biological annotations and demonstrate its use for biological discovery in a meta-analysis of human embryonic development. We integrate a CellWhisperer chat box with the CELLxGENE browser, allowing users to interactively explore gene expression through a combined graphical and chat interface. In summary, CellWhisperer leverages large community-scale data repositories to connect transcriptomes and text, thereby enabling interactive exploration of single-cell RNA-sequencing data with natural-language chats. Gene regulation in immune cells Machine learning Preclinical research Software Transcriptomics
N Nature Biotechnology · Nov 10, 2025 Clearance of intracranial debris by ultrasound reduces inflammation and improves outcomes in hemorrhagic stroke models Impaired clearance of neurotoxic debris in the brain exacerbates neurologic disease and presents a promising therapeutic target. Pharmacologic therapies can enhance meningeal lymphatic clearance in preclinical models but may be limited by systemic toxicities or invasive administration. Here we report a low-intensity, focused ultrasound protocol that noninvasively clears pathogenic substances from the cerebrospinal fluid and brain interstitium in mice. Using two models of hemorrhagic stroke, we demonstrate that this protocol clears the cerebrospinal fluid and interstitium of blood cells, which accumulate in the deep cervical lymph nodes via meningeal lymphatics. The protocol directly modulates molecular processes, including mechanosensitive channels, to shift microglial phenotypes and astrocytic aquaporin localization to reduce neuroinflammation and neurocytotoxicity. In the intracerebral hemorrhage model, it improves behavioral outcomes and increases survival with greater efficacy than a pharmacologic benchmark. The protocol satisfies Food and Drug Administration safety guidelines, supporting clinical translatability. If demonstrated effective clinically, it may provide therapeutic benefit not only in hemorrhagic stroke but also in other neurologic disorders that involve impaired debris clearance. Biomedical engineering Stroke
N Nature Biotechnology · Nov 07, 2025 RNA stability enhancers for durable base-modified mRNA therapeutics The limited stability of mRNA in vivo remains a major challenge for vaccines and therapeutics. While alternative RNA formats such as circular RNA or self-amplifying RNA offer greater durability, these modalities often suffer from low translation, modification incompatibility and difficult manufacturing. To overcome these limitations, we screen 196,277 viral sequences and identify eleven elements that strongly enhance mRNA stability and translation. Mechanistically, they recruit TENT4 to extend the poly(A) tail, preventing deadenylation. Five of them are compatible withN1-methylpseudouridine, which improves mRNA efficacy and reduces immunogenicity. An element named A7 demonstrates particularly robust performance across cell types, delivery methods, modifications and coding sequences, making linear mRNA as stable as circular RNA while achieving higher translation efficiency. In mouse liver, A7-containing linear mRNA exhibits substantially higher protein levels than circular RNA, with sustained expression lasting for over 2 weeks. These RNA stability enhancers enable robust linear mRNA platforms that combine high and durable expression, low immunogenicity and simple manufacturing. Molecular biology Nucleic-acid therapeutics
N Nature Biotechnology · Nov 06, 2025 CRISPR live-cell imaging reveals chromatin dynamics and enhancer interactions at multiple non-repetitive loci Existing methods to visualize dynamic changes in the three-dimensional genome, promoter−enhancer interactions and the influence of epigenetic modifications in non-repetitive loci are limited. Here we introduce CRISPR PRO-LiveFISH (Pooled gRNAs with Orthogonal bases LiveFISH), which combines orthogonal bases from expanded genetic alphabet technology and rational single guide RNA (sgRNA) design to efficiently label multiple non-repetitive loci in living cells. The optimized method allows simultaneous imaging of up to six genomic loci and uses as few as 10 sgRNAs for non-repetitive loci imaging without signal amplification. We demonstrate the method in diverse cell types, including primary cells, and apply it to reveal enhancer−promoter dynamics and a correlation between genomic dynamics and epigenetic states. We also show thatPCDHα−enhancer interactions may persist despite spatial mobility and that BRD4 maintains super-enhancer contacts regulatingMYConcogene expression in cancer cells. CRISPR PRO-LiveFISH can be applied to diverse studies of chromatin dynamics and genome organization in living cells. Biotechnology Cellular imaging Genetics
N Nature Biotechnology · Nov 06, 2025 Site-specific DNA insertion into the human genome with engineered recombinases Insertions of large DNA sequences into the genome are broadly enabling for research and therapeutic applications. Large serine recombinases (LSRs) can mediate direct, site-specific genomic integration of multi-kilobase DNA sequences without a pre-installed landing pad, albeit with low insertion rates and high off-target activity. Here we present an engineering roadmap for jointly optimizing their DNA recombination efficiency and specificity. We combine directed evolution, structural analysis and computational models to rapidly identify additive mutational combinations. We further enhance performance through donor DNA optimization and dCas9 fusions, enabling simultaneous target and donor recruitment. Our top engineered LSR variants, superDn29−dCas9, goldDn29−dCas9 and hifiDn29−dCas9, achieve up to 53% integration efficiency and 97% genome-wide specificity at an endogenous human locus and effectively integrate large DNA cargoes up to 12 kb for stable expression in non-dividing cells, stem cells and primary human T cells. Rational engineering of DNA recombinases enables precise and efficient single-step genome insertion for diverse applications across gene and cell therapies. DNA recombination Gene therapy Protein design
N Nature Biotechnology · Nov 05, 2025 A nonsurgical brain implant enabled through a cell–electronics hybrid for focal neuromodulation Bioelectronic implants for brain stimulation are used to treat brain disorders but require invasive surgery. To provide a noninvasive alternative, we report nonsurgical implants consisting of immune cell–electronics hybrids, an approach we call Circulatronics. The devices can be delivered intravenously and traffic autonomously to regions of inflammation in the brain, where they implant and enable neuromodulation, circumventing the need for surgery. To achieve suitable electronics, we designed and built subcellular-sized, wireless, photovoltaic electronic devices that harvest optical energy with high power conversion efficiency. In mice, we demonstrate nonsurgical implantation in an inflamed brain region, as an example of therapeutic target for several neural diseases, by employing monocytes as cells, covalently attaching them to the subcellular-sized, wireless, photovoltaic electronic devices and administering the resulting hybrids intravenously. We also demonstrate neural stimulation with 30-µm precision around the inflamed region. Thus, by fusing electronic functionality with the biological transport and targeting capabilities of living cells, this technology can form the foundation for autonomously implanting bioelectronics. Biomedical engineering Biotechnology
N Nature Biotechnology · Nov 04, 2025 KATMAP infers splicing factor activity and regulatory targets from knockdown data Typical RNA sequencing (RNA-seq) experiments uncover hundreds of splicing changes, reflecting underlying changes in splicing factor (SF) activity. Understanding how SF activity influences transcriptomic variation requires elucidating how each SF impacts splicing. Here, we present an interpretable regression model, KATMAP, which models splicing changes throughout the transcriptome by analyzing changes in SF binding and the resulting alterations in RNA processing. To learn a regulatory model, KATMAP requires SF perturbation RNA-seq data and the SF’s binding motif as inputs, returning a description of the SF’s position-specific regulatory activity and predicted targets. The KATMAP software includes models pretrained on ENCODE SF knockdown data. Learned KATMAP models can be applied to predict SF regulation andcis-elements at individual exons, which can guide the design of splice-switching antisense oligonucleotides. KATMAP can also interpret RNA-seq data by uncovering the factors responsible for transcriptomic changes, distinguishing direct SF targets from indirect effects and inferring relevant SFs from clinical RNA-seq data. Computational models Software Transcriptomics
N Nature Biotechnology · Nov 03, 2025 Growth-coupled microbial biosynthesis of the animal pigment xanthommatin Engineering heterologous natural product pathways in bacteria has achieved broad success but most approaches suffer from low initial production levels that require extensive, resource-heavy iterative strain optimization. Xanthommatin is a structurally complex, color-changing animal ommochrome with material and cosmetic applications, yet production in microbial cell factories has been difficult. Here, we introduce a growth-coupled biosynthetic strategy involving a feedback loop where an excised one-carbon (C1) moiety is used as a driver of bacterial growth, simultaneously boosting bioproduction of the target compound. This broadly applicable, plug-and-play strategy is illustrated by enabling xanthommatin biosynthesis in a 5,10-methylenetetrahydrofolate auxotroph of the platform soil bacteriumPseudomonasputida. In this design, formate released during xanthommatin production relieves the C1 deficiency, thereby effectively coupling bacterial growth to pigment synthesis. Adaptive laboratory evolution streamlined xanthommatin’s gram-scale bioproduction from glucose, establishing C1 restoration as a general biosynthetic approach to accelerate the engineering of natural product biosynthesis in bacteria. Biosynthesis Metabolic engineering Small molecules
N Nature Biotechnology · Oct 30, 2025 High-plex spatial RNA imaging in one round with conventional microscopes using color-intensity barcodes Spatial RNA imaging has not been widely adopted because conventional fluorescence microscopy is limited to only a few channels and the cyclic reactions needed to increase multiplexing in techniques such as sequential fluorescence in situ hybridization require sophisticated instrumentation. Here, we introduce ‘profiling of RNA in situ through single-round imaging’ (PRISM), a method that expands coding capacity through color intensity grading. Using a radius vector filtering strategy to ensure the distinguishability of codewords in color space, PRISM achieves up to 64-plex color-barcoded RNA imaging in a single imaging round with conventional microscopes. We validate PRISM’s versatility across various tissues by generating a three-dimensional (3D) atlas of mouse embryonic development from E12.5 to E14.5, a quasi-3D tumor–normal transition landscape of human hepatocellular carcinoma and a 3D cell atlas and subcellular RNA localization landscapes of mouse brain. Additionally, we show the critical role of cancer-associated fibroblasts in mediating immune infiltration and immune response heterogeneity within and between tumor microenvironments. Fluorescence imaging Fluorescence in situ hybridization Transcriptomics
N Nature Biotechnology · Oct 30, 2025 Mesenchymal thymic niche cells enable regeneration of the adult thymus and T cell immunity Thymic atrophy and the progressive immune decline that accompanies it is a major health problem, chronically with age and acutely with immune injury. No definitive solution is available. Here we demonstrate that one of the three mesenchymal cell subsets identified by single-cell analysis of human and mouse thymic stroma is a critical niche component for T lymphopoiesis. The Postn+subset is perivascular, and its depletion abrogates T cell progenitor recruitment, likely through production of the chemokine Ccl19. It markedly declines with age and in the acute setting of hematopoietic stem cell transplant conditioning. When isolated and adoptively transferred, Postn+cells durably engraft the atrophic thymus, recruit early T progenitors, increase T cell neogenesis and enhance T cell response to vaccination. More readily available mesenchymal populations expressing Ccl19 provide similar effects. These data define a thymus lymphopoietic niche cell type that may be manipulated therapeutically to regenerate T lymphopoiesis. Bone marrow transplantation Lymphopoiesis
N Nature Biotechnology · Oct 28, 2025 Selective RNA sequestration in biomolecular condensates directs cell fate transitions Controlling stem cell differentiation is a longstanding goal in biomedical research. Here we explore how cell fate is influenced by RNA condensates, specifically P-bodies, which modulate gene expression posttranscriptionally. We profiled the transcriptomes of biomolecular condensates in diverse developmental contexts spanning multiple vertebrate species. Our analyses revealed conserved, cell type-specific sequestration of untranslated RNAs encoding cell fate regulators. P-body RNA contents do not reflect active gene expression in each cell type but are enriched for translationally repressed transcripts characteristic of the preceding developmental stage. Mechanistically, P-body contents are controlled by microRNAs and can be profoundly reshaped by perturbing AGO2 or polyadenylation site usage. Applying these insights to stem cell differentiation, we show that manipulating P-body assembly or microRNA activity can direct naive mouse and human pluripotent stem cells toward totipotency or primed human embryonic cells toward the germ cell lineage. Our findings link cell fate decisions to RNA condensates across vertebrates and provide a means of controlling cell identity. Embryonic stem cells RNA metabolism Totipotent stem cells
N Nature Biotechnology · Oct 28, 2025 A single-chain derivative of an integrin-activating antibody potentiates organoid growth in Matrigel and collagen hydrogels Current methods of culturing human epithelial organoids from adult stem cells may not be compatible with clinical applications as they rely on xenogeneic, chemically undefined or non-standardized components such as the basement membrane extract Matrigel. Matrigel provides a source of extracellular matrix molecules, including laminins and collagen IV, which interact with β1 integrins expressed on organoid cells. Here we describe a single-chain (sc) version of antibody TS2/16 that allosterically activates integrin β1 function in organoids. The addition of monomeric scTS2/16 to organoid medium results in up to a fivefold increase in the yield of all gastrointestinal organoids grown in Matrigel. Moreover, scTS2/16 supports a six- to sevenfold increase in the yield of these organoids when cultured in collagen I hydrogels, both in 3D and 2D. Collagen I is well defined, available in clinical-grade formulations and, when combined with scTS2/16, may support the clinical application of epithelial organoids derived from gastrointestinal tissues and other epithelial sources. Biotechnology Medical research
N Nature Biotechnology · Oct 24, 2025 Democratizing protein language model training, sharing and collaboration Training and deploying large-scale protein language models typically requires deep machine learning expertise—a barrier for researchers outside this field. SaprotHub overcomes this challenge by offering an intuitive platform that facilitates training and prediction as well as storage and sharing of models. Here we provide the ColabSaprot framework built on Google Colab, which potentially powers hundreds of protein training and prediction applications, enabling researchers to collaboratively build and share customized models. Bioinformatics Protein function predictions
N Nature Biotechnology · Oct 24, 2025 Deep-learning-based virtual screening of antibacterial compounds The increase in multidrug-resistant bacteria underscores an urgent need for additional antibiotics. Here, we integrate small-molecule high-throughput screening with a deep-learning-based virtual screening approach to uncover new antibacterial compounds. We screen ~2 million small molecules against a sensitizedEscherichia colistrain, yielding thousands of hits. We use these data to train a deep learning model, GNEprop, to predict antibacterial activity, retrospectively validating robustness with respect to out-of-distribution generalization and activity cliff prediction. Virtual screening of over 1.4 billion synthetically accessible compounds identifies potential candidates, of which 82 exhibit antibacterial activity on the same strain, illustrating a 90-fold improved hit rate over the high-throughput screening experiment used for training. Many newly identified compounds exhibit high dissimilarity to known antibiotics, potency beyond the training bacterial strain and selectivity. Biological characterization identifies specific, validated targets, indicating promising avenues for further exploration in antibiotic discovery. High-throughput screening Machine learning Virtual drug screening Virtual screening
N Nature Biotechnology · Oct 23, 2025 Discovery and engineering of retrons for precise genome editing Retrons can produce multicopy single-stranded DNA in cells through self-primed reverse transcription. However, their potential for inserting genetic cargos in eukaryotes remains largely unexplored. Here we report the discovery and engineering of highly efficient retron-based gene editors for mammalian cells and vertebrates. Through bioinformatic analysis of metagenomic data and functional screening, we identify retron reverse transcriptases that are highly active in mammalian cells. Rational design further improves the editing efficiency to levels comparable with conventional single-stranded oligodeoxynucleotide donors but from a genetically encoded cassette. Retron editors exhibit robust activity with Cas12a nuclease and Cas9 nickase, expanding the genomic target scope and bypassing the need for a DNA double-stranded break. Using a rationally engineered retron editor, we incorporate a split GFP epitope tag for live-cell imaging. Lastly, we develop an all-RNA delivery strategy to enable DNA-free gene editing in cells and vertebrate embryos. This work establishes retron editors as a versatile and efficient tool for precise gene editing. DNA Targeted gene repair
N Nature Biotechnology · Oct 23, 2025 Live imaging of late-stage preimplantation human embryos reveals de novo mitotic errors Existing methods to image chromosome segregation errors are not suitable for studying human embryos at advanced preimplantation stages. As chromosomal errors are a leading cause of miscarriage and infertility, it remains unclear whether missegregation arises postfertilization. Here we optimize nuclear DNA labeling via messenger RNA electroporation and apply light-sheet live imaging to reveal chromosome segregation errors immediately before implantation. We show that embryos at advanced preimplantation stages display missegregation, including multipolar spindle formation, lagging chromosomes, misalignment and mitotic slippage. Most lagging chromosomes are passively inherited rather than reincorporated. To trace individual nuclei, we developed an open-source, semi-automated segmentation method using a customized deep learning model optimized for variability in embryo size, shape and signal. With this approach, we find most labeled cells remain externally positioned, consistent with placental rather than inner cell mass fate. Our findings raise questions about clinical uses of preimplantation genetic testing for aneuploidy, while providing broadly applicable imaging and segmentation methods for studying diverse cellular structures in human embryos. Chromosome segregation Image processing Time-lapse imaging
N Nature Biotechnology · Oct 23, 2025 RNA structure modulates Cas13 activity and enables mismatch detection Cas13 is activated by the hybridization of a CRISPR RNA to a complementary single-stranded RNA protospacer in a target RNA. While Cas13 is not activated by double-stranded RNA in vitro, it robustly targets RNA in cellular environments where RNAs are highly structured. The mechanism by which Cas13 targets structured RNAs remains unknown. Here, we systematically probe the effects of secondary structure on Cas13. We find that secondary structure in the protospacer and 3′ to it inhibits Cas13 activity and quantitatively explains the former effect through a strand displacement framework. We then harness strand displacement to generate an ‘occluded’ Cas13, which enhances mismatch discrimination up to 50-fold and enables sequence-agnostic mutation identification at low (<1%) allele frequencies. Using occluded Cas13, we identify human-adaptive mutations in SARS-CoV-2 and human and avian influenza A viruses, as well as oncogenic mutations inKRAS. Our work leverages improved mechanistic understanding of Cas13 to expand the scope of RNA diagnostics and enable structure-informed Cas13 approaches. Enzyme mechanisms Infectious-disease diagnostics Influenza virus Molecular engineering RNA
N Nature Biotechnology · Oct 23, 2025 Elucidating lipid nanoparticle properties and structure through biophysical analyses Designing lipid nanoparticle (LNP) delivery systems with specific targeting, potency and minimal side effects is crucial for their clinical use. However, traditional characterization methods, such as dynamic light scattering, cannot accurately quantify physicochemical properties of LNPs and how these are influenced by the lipid composition and mixing method. Here, we structurally characterize polydisperse LNP formulations by applying emerging solution-based biophysical methods that have higher resolution and provide biophysical data beyond size and polydispersity. These techniques include sedimentation velocity analytical ultracentrifugation, field-flow fractionation followed by multiangle light scattering and size-exclusion chromatography in line with synchrotron small-angle X-ray scattering. We show that LNPs have intrinsic polydispersity in size, RNA loading and shape, which depend on both the formulation technique and the lipid composition. Lastly, we predict LNP transfection in vitro and in vivo by examining the relationship between mRNA translation and physicochemical characteristics. Solution-based biophysical methods will be essential for determining LNP structure–function relationships, facilitating the creation of new design rules for LNPs. Drug delivery Nanoparticles Nanoscale biophysics
N Nature Biotechnology · Oct 22, 2025 Tissue and cellular spatiotemporal dynamics in colon aging Tissue structure and molecular circuitry in the colon can be profoundly impacted by systemic age-related effects but many of the underlying molecular cues remain unclear. Here, we build a cellular and spatial atlas of the colon across three anatomical regions and 11 age groups, encompassing ~1,500 mouse gut tissues profiled by spatial transcriptomics and ~400,000 single nucleus RNA-sequencing profiles. We develop a computational framework, cSplotch, which learns a hierarchical Bayesian model of spatially resolved cellular expression associated with age, tissue region and sex by leveraging histological features to share information across tissue samples and data modalities. Using this model, we identify cellular and molecular gradients along the adult colonic tract and across the main crypt axis and multicellular programs associated with aging in the large intestine. Our multimodal framework for the investigation of cell and tissue organization can aid in the understanding of cellular roles in tissue-level pathology. Ageing Bayesian inference Data integration Gene expression analysis RNA sequencing
N Nature Biotechnology · Oct 21, 2025 Integrated epigenetic and genetic programming of primary human T cells Targeted epigenetic engineering of gene expression in cell therapies would allow programming of desirable phenotypes without many of the challenges and safety risks associated with double-strand break-based genetic editing approaches. Here, we develop an all-RNA platform for efficient, durable and multiplexed epigenetic programming in primary human T cells, stably turning endogenous genes off or on using CRISPRoff and CRISPRon epigenetic editors. We achieve epigenetic programming of diverse targeted genomic elements without the need for sustained expression of CRISPR systems. CRISPRoff-mediated gene silencing is maintained through numerous cell divisions, T cell stimulations and in vivo adoptive transfer, avoiding cytotoxicity or chromosomal abnormalities inherent to multiplexed Cas9-mediated genome editing. Lastly, we successfully combined genetic and epigenetic engineering using orthogonal CRISPR Cas12a–dCas9 systems for targeted chimeric antigen receptor (CAR) knock-in and CRISPRoff silencing of therapeutically relevant genes to improve preclinical CAR-T cell-mediated in vivo tumor control and survival. Immunology Synthetic biology
N Nature Biotechnology · Oct 21, 2025 AlphaDIA enables DIA transfer learning for feature-free proteomics The scale of data generated for mass-spectrometry-based proteomics and modern acquisition strategies poses a challenge to bioinformatic analysis. Search engines need to make optimal use of the data for biological discoveries while remaining statistically rigorous, transparent and performant. Here we present alphaDIA, a modular open-source search framework for data-independent acquisition (DIA) proteomics. We developed a feature-free identification algorithm that performs machine learning directly on the raw signal and is particularly suited for detecting patterns in data produced by time-of-flight instruments. Benchmarking demonstrates competitive identification and quantification performance. While the method supports empirical spectral libraries, we propose a search strategy named DIA transfer learning that uses fully predicted libraries. This entails continuously optimizing a deep neural network for predicting machine-specific and experiment-specific properties, enabling the generic DIA analysis of any post-translational modification. AlphaDIA provides a high performance and accessible framework running locally or in the cloud, opening DIA analysis to the community. Proteome informatics Proteomics
N Nature Biotechnology · Oct 17, 2025 A tumor-on-a-chip for in vitro study of CAR-T cell immunotherapy in solid tumors Our limited understanding of cancer–immune interactions remains a critical barrier to advancing chimeric antigen receptor (CAR)-T cell therapy for solid malignancies. Here, we present a microengineered system that enables vascularization of human tumor explants and their controlled perfusion with immune cells to model the activity of CAR-T cells in the tumor microenvironment. Using vascularized human lung adenocarcinoma tumors, we first demonstrate the ability of our tumor-on-a-chip system to simulate, visualize and interrogate CAR-T cell function. We then test a chemokine-directed CAR-T cell engineering strategy in a model of malignant pleural mesothelioma and validate our findings in a matching in vivo mouse model. Finally, we describe a potential therapeutic target that can be pharmacologically modulated to increase the efficacy of CAR-T cells in lung adenocarcinoma, for which we present biomarkers identified by global metabolomics analysis. Our microphysiological system provides promising in vitro technology to advance the development of adoptive cell therapies for cancer and other diseases. Biomedical engineering Cancer immunotherapy Cancer models Lab-on-a-chip Tissue engineering
N Nature Biotechnology · Oct 16, 2025 Accurate somatic small variant discovery for multiple sequencing technologies with DeepSomatic Somatic variant detection is an integral part of cancer genomics analysis. While most methods have focused on short-read sequencing, long-read technologies offer potential advantages in repeat mapping and variant phasing. We present DeepSomatic, a deep-learning method for detecting somatic small nucleotide variations and insertions and deletions from both short-read and long-read data. The method has modes for whole-genome and whole-exome sequencing and can run on tumor–normal, tumor-only and formalin-fixed paraffin-embedded samples. To train DeepSomatic and help address the dearth of publicly available training and benchmarking data for somatic variant detection, we generated and make openly available the Cancer Standards Long-read Evaluation (CASTLE) dataset of six matched tumor–normal cell line pairs whole-genome sequenced with Illumina, PacBio HiFi and Oxford Nanopore Technologies, along with benchmark variant sets. Across samples, both cell line and patient-derived, and across short-read and long-read sequencing technologies, DeepSomatic consistently outperforms existing callers. Genome informatics Machine learning
N Nature Biotechnology · Oct 15, 2025 Predicting functions of uncharacterized gene products from microbial communities The majority of genes in microbial communities remain uncharacterized. Here we develop a method to infer putative function for microbial proteins at scale by assessing community-wide multiomics data. We predict high-confidence functions for >443,000 protein families (~82.3% previously uncharacterized), including >27,000 protein families with weak homology to known proteins and >6,000 protein families without homology. These were drawn from 1,595 gut metagenomes and 800 metatranscriptomes from the Integrative Human Microbiome Project (HMP2/iHMP). Integrating additional information such as sequence similarity, genomic proximity and domain–domain interactions improves performance of the method. Our method’s implementation, FUGAsseM, is generalizable and predicts protein function in both well-studied and undercharacterized communities. FUGAsseM achieves similar levels of accuracy in the context of microbial communities when compared to state-of-the-art approaches designed for application to single organisms while simultaneously providing much greater breadth of coverage. This initial study expands the functional landscape of the human gut microbiome and allows for exploration of microbial proteins in undercharacterized communities. Data integration Gene expression Microbiome Protein function predictions Software
N Nature Biotechnology · Oct 13, 2025 Engineered pattern recognition receptors enhance broad-spectrum plant resistance Conventional plant resistance breeding has primarily focused on intracellular immune receptors, while cell-surface pattern recognition receptors (PRRs) have been underexplored because of their comparatively modest contributions to resistance. However, PRRs offer untapped potential for crop improvement. Here we demonstrate that theArabidopsisreceptor-like protein RLP23, which recognizes molecular patterns from three distinct microbial kingdoms, confers broad-spectrum resistance when introduced into the Solanaceae crop tomato. We also identify the C-terminal domain of RLPs as crucial for ensuring compatibility and efficacy during heterologous expression. Targeted engineering of the C-terminal domain significantly enhances RLP23 and its use, enabling transfer of robust resistance against bacterial, fungal and oomycete pathogens to tomato plants without compromising yield. We extend this RLP engineering strategy to rice and poplar, highlighting its broad applicability. These findings establish a versatile framework for PRR-based engineering, opening additional avenues for sustainable crop protection. Molecular engineering in plants Pattern recognition receptors in plants
N Nature Biotechnology · Oct 13, 2025 Programmable promoter editing for precise control of transgene expression Subtle changes in gene expression direct cells to distinct cellular states. Identifying and controlling dose-dependent transgenes require tools for precisely titrating expression. Here, we develop a highly modular, extensible framework called DIAL for building editable promoters that allow for fine-scale, heritable changes in transgene expression. Using DIAL, we increase expression by recombinase-mediated excision of spacers between the binding sites of a synthetic zinc finger transcription factor and the core promoter. By nesting varying numbers and lengths of spacers, DIAL generates a tunable range of unimodal setpoints from a single promoter. Through small-molecule control of transcription factors and recombinases, DIAL supports temporally defined, user-guided control of transgene expression that is extensible to additional transcription factors. Lentiviral delivery of DIAL generates multiple setpoints in primary cells and induced pluripotent stem cells. As promoter editing generates stable states, DIAL setpoints are heritable, facilitating mapping of transgene levels to phenotype and fate in direct conversion to induced motor neurons. The DIAL framework opens opportunities for tailoring transgene expression and improving the predictability and performance of gene circuits across diverse applications. Stem-cell biotechnology Synthetic biology Transcriptional regulatory elements
N Nature Biotechnology · Oct 07, 2025 Folding-mediated secretion of pure bispecific antibodies Bispecific antibodies (bsAbs) can enable therapeutic mechanisms, such as dual antigen targeting or receptor agonism, that are impossible using monoclonal antibodies. BsAbs with IgG-like format (bsIgG) are comprised of two unique heavy chains, each having a cognate light chain. Co-expression of these four unique polypeptides often leads to several mispaired species that are difficult to separate from the target bsIgG due to their similar biophysical properties. Here we describe a set of mutations called ProAla that exploit a the unfolded protein response pathway of cells. ProAla heavy chains are engineered with higher folding energy barriers such that only the cognate light and heavy chains can induce folding, chaperone release and secretion. The structures of the ProAla Fab and Fc regions are identical in structure to normal antibodies, enabling maintenance of half-life and function. Mispaired polypeptides fail to secrete from the cell due to enhanced interaction with the endoplasmic reticulum chaperone BiP, resulting in increased purity of secreted bsIgGs. Antibody therapy Endoplasmic reticulum
N Nature Biotechnology · Oct 07, 2025 Generation of modified cows and sheep from spermatid-like haploid embryonic stem cells In rodents, injection of haploid androgenetic embryonic stem cells (haES cells) into intact oocytes enables full-term development of offspring. The value of this method in research and genome engineering has not been replicated in ruminants because ruminant haES cells are yet to be obtained. Here we report the derivation of cow and sheep haES cells and their application to generate offspring by a method we call intracytoplasmic haES cell injection (iCHI), in analogy with intracytoplasmic sperm injection. Ruminant haES cells display characteristics of formative-state pluripotency and differentiate into the three germ layers both in vitro and in vivo. Ectopic expression of protamine in haES cells converts their nuclei into a spermatid-like structure and improves full-term development of iCHI embryos, a method we call protamine iCHI (Pro-iCHI). We also combine Pro-iCHI and prime editing to generate gene-modified cows and sheep. Overall, Pro-iCHI provides a promising approach for production of genetically modified livestock. Animal breeding Cloning Embryonic stem cells Genetic engineering Stem-cell biotechnology
N Nature Biotechnology · Oct 02, 2025 A trimodal protein language model enables advanced protein searches ProTrek unifies protein sequence, structure and natural language function in a trimodal language model through contrastive learning, enabling comprehensive searches between any two modalities, including within modality. ProTrek surpasses current alignment tools (for example, Foldseek and MMseqs2) in speed and accuracy for identifying functionally related proteins. Computational and wet-lab experimental validations show that the ProTrek server (www.search-protrek.com), with precomputed embeddings for over 5 billion proteins, efficiently processes and analyzes large-scale protein repositories. Bioinformatics Genetic engineering Molecular evolution Protein databases Sequence annotation
N Nature Biotechnology · Oct 02, 2025 Discovery and protein language model-guided design of hyperactive transposases The diversity and biochemical potential of thePiggyBactransposase gene insertion system remains largely unexplored. Using a eukaryotic transposon mining pipeline, we expand the explored diversity by two orders of magnitude and experimentally validate a subset of highly divergentPiggyBacsequences. Fine-tuning a protein language model to further expandPiggyBacsequence space discovers transposases with improved activity and that are compatible with T cell engineering and Cas9-directed transposase-assisted integration. Evolutionary biology Genetic engineering Genetic vectors
N Nature Biotechnology · Oct 01, 2025 Detecting and quantifying circular RNAs in terabyte-scale RNA-seq datasets with CIRI3 To address recent challenges in circular RNA (circRNA) analysis, we present CIRI3, a tool for circRNA detection and quantification in terabyte-scale RNA-sequencing datasets. Using dynamic multithreaded task partitioning and a blocking search strategy for junction reads, CIRI3 is an order of magnitude faster than existing tools, while providing increased accuracy. We identified differentially spliced circRNAs across 2,535 cancer-related samples, and constructed a pretraining model and a biomarker network provided as the CIRIonco database. Data mining Gene expression
N Nature Biotechnology · Oct 01, 2025 Design of optimized epigenetic regulators for durable gene silencing with application toPCSK9in nonhuman primates Epigenetic editing is a promising strategy for modifying gene expression while avoiding the permanent alterations and potential genotoxicity of genome-editing technologies. Here we designed optimized epigenetic regulators (EpiRegs) by testing combinations of transcription activator-like effector (TALE)-based and catalytically deactivated Cas9 (dCas9)-based epigenetic modification effectors and fusion protein structures. TALE-based EpiReg (EpiReg-T) achieved a final efficiency of 98% in mice, surpassing the initial dCas9-based efficiency of 64%. We demonstrated the approach in macaques by introducing DNA methylation and histone modifications to inhibit proprotein convertase subtilisin/kexin type 9 (PCSK9) expression, thereby lowering low-density lipoprotein cholesterol levels. A single dose of EpiReg-T delivered with lipid nanoparticles achieved efficient (>90%) and long-lasting (343 days) silencing ofPCSK9in the liver. Integrative multiomic analyses revealed minimal off-target effects in EpiReg-T-treated monkeys, mice and human-derived cells. EpiReg can be redirected to other genes by reengineering the DNA-binding domain. Our findings represent a step toward the clinical application of epigenetic editing for the treatment of human diseases. Drug development Gene therapy
N Nature Biotechnology · Oct 01, 2025 Scaling DNA synthesis with a microchip-based massively parallel synthesis system Current high-throughput DNA synthesis technologies use intricate chip and microfluidic systems to produce large-scale synthetic oligonucleotides but with low concentration and limited compatibility for long DNA assembly. Here we report a massive-in-parallel synthesis system, with an ‘identification–sorting–synthesis–recycling’ iteration mechanism applied to microchips for high-throughput DNA synthesis. This approach increases DNA product concentration by four to six orders of magnitude and simplifies downstream processes for large-scale gene synthesis. Synthetic biology
N Nature Biotechnology · Sep 30, 2025 Derivation of embryonic stem cells across avian species Germline-competent embryonic stem (ES) cells have been successfully derived from mice and rats, but not from other species. Here we report the development of culture conditions for deriving ES cells from chickens and seven other avian species. Chicken ES cells express core pluripotency markers and can differentiate into cells of all embryonic germ layers, as well as extra-embryonic lineages. Notably, chicken ES cells contribute to high rates of chimerism when injected into chicken embryos and give rise to germ cells both in vitro and in ovo, confirming their germline competence. In addition, we demonstrated that ES cell self-renewal pathways are conserved among avian species, allowing ES cells from multiple avian species to be established using optimized chicken ES cell culture conditions. The establishment of authentic avian ES cells lays the groundwork for future applications in genetic engineering and the conservation of avian biodiversity. Embryonic stem cells Genetic engineering
N Nature Biotechnology · Sep 12, 2025 Bioactive molecules unearthed by terabase-scale long-read sequencing of a soil metagenome Metagenomics provides access to the genetic diversity of uncultured bacteria through analysis of DNA extracted from whole microbial communities. Long-read sequencing is advancing metagenomic discovery by generating larger DNA assemblies than previously possible. However, harnessing the potential of long-read sequencing to access the vast diversity within soil microbiomes is hampered by the challenge of isolating high-quality DNA. Here we introduce a method that can liberate large, high-quality metagenomic DNA fragments from soil bacteria and pair them with optimized nanopore long-read sequencing to generate megabase-sized assemblies. Using this method, we uncover hundreds of complete circular metagenomic genomes from a single soil sample. Through a combination of bioinformatic prediction and chemical synthesis, we convert nonribosomal peptide biosynthetic gene clusters directly into bioactive molecules, identifying antibiotics with rare modes of action and activity against multidrug-resistant pathogens. Our approach advances metagenomic access to the vast genetic diversity of the uncultured bacterial majority and provides a means to convert it to bioactive molecules. Antibiotics Soil microbiology
N Nature Biotechnology · Sep 10, 2025 A rapid imaging-based screen for induced-proximity degraders identifies a potent degrader of oncoprotein SKP2 Targeted protein degraders hold potential as therapeutic agents to target conventionally ‘undruggable’ proteins. Here, we develop a high-throughput screen, DEath FUSion Escaper (DEFUSE), to identify small-molecule protein degraders. By conjugating the protein of interest to a fast-acting triggerable death protein, this approach translates target protein degradation into a cell survival phenotype to illustrate the presence of degraders. Using this method, we discovered a small molecule (SKPer1) that triggers degradation of the oncoprotein SKP2 and specifically kills SKP2-expressing cancer cells. Mechanistically, SKPer1 acts as an induced-proximity degrader by inducing interaction between SKP2 and an E3 ligase, STUB1, resulting in SKP2 ubiquitination and degradation. SKPer1 exhibits substantial tumour suppression with good safety profiles in vivo. We further show that a sequence of ten amino acids from SKP2 can serve as a versatile degradation tag. Cancer Drug discovery
N Nature Biotechnology · Sep 10, 2025 Efficient sequence alignment against millions of prokaryotic genomes with LexicMap The size of microbial sequence databases continues to grow beyond the abilities of existing alignment tools. We introduce LexicMap, a nucleotide sequence alignment tool for efficiently querying moderate-length sequences (>250 bp) such as a gene, plasmid or long read against up to millions of prokaryotic genomes. We construct a small set of probek-mers, which are selected to efficiently sample the entire database to be indexed such that every 250-bp window of each database genome contains multiple seedk-mers, each with a shared prefix with one of the probes. Storing these seeds in a hierarchical index enables fast and low-memory alignment. We benchmark both accuracy and potential to scale to databases of millions of bacterial genomes, showing that LexicMap achieves comparable accuracy to state-of-the-art methods but with greater speed and lower memory use. Our method supports querying at scale and within minutes, which will be useful for many biological applications across epidemiology, ecology and evolution. Bacterial genomics Computational models Genetic databases Genome informatics Software
N Nature Biotechnology · Sep 09, 2025 Genome-wide mapping of RNA-protein associations through sequencing RNA-protein interactions critically regulate gene expression and cellular processes, yet their comprehensive mapping remains challenging due to their structural diversity. We introduce PRIM-seq (protein-RNA interaction mapping by sequencing), a method for concurrent de novo identification of RNA-binding proteins and their associated RNAs. PRIM-seq generates unique chimeric DNA sequences by proximity ligation of RNAs with protein-linked DNA barcodes, which are subsequently decoded through sequencing. We apply PRIM-seq to two human cell lines and construct a human RNA-protein association network (HuRPA), encompassing >350,000 associations involving ~7,000 RNAs and ~11,000 proteins, including 2,610 proteins that each interact with at least 10 distinct RNAs. We experimentally validate the tumorigenesis-associated lincRNA LINC00339, the RNA with the highest number of protein associations in HuRPA, as a protein-associated RNA. We further validate the RNA-associating abilities of chromatin-conformation regulators SMC1A, SMC3 and RAD21, as well as the metabolic enzyme PHGDH. PRIM-seq enables systematic discovery and prioritization of RNA-binding proteins and their targets without gene- or protein-specific reagents. Bioinformatics Biotechnology
N Nature Biotechnology · Sep 09, 2025 Antibody–bottlebrush prodrug conjugates for targeted cancer therapy Antibody–drug conjugates (ADCs) are effective targeted therapeutics but are limited in their ability to incorporate less-potent payloads, varied drug mechanisms of action, different drug release mechanisms and tunable drug-to-antibody ratios. Here we introduce a technology to overcome these limitations called ‘antibody–bottlebrush prodrug conjugates’ (ABCs). An ABC consists of an IgG1 monoclonal antibody covalently conjugated to the terminus of a compact bivalent bottlebrush prodrug that has payloads bound through cleavable linkers and polyethylene glycol branches. This design enables the synthesis of ABCs with tunable average drug-to-antibody ratios up to two orders of magnitude greater than those of traditional ADCs. We demonstrate the functional flexibility and manufacturing efficiency of this technology by synthesizing more than 10 different ABCs targeting either HER2 or MUC1 using drugs with potencies spanning several orders of magnitude as well as imaging agents for ABC visualization and photocatalysts for proximity-based labeling of the ABC interactome. ABCs display high target engagement, high cell uptake and improved efficacy in tumor models compared to conventional HER2-targeted ADCs, suggesting promise for clinical translation. Drug delivery Polymer synthesis
N Nature Biotechnology · Sep 04, 2025 Targeted DNA ADP-ribosylation triggers templated repair in bacteria and base mutagenesis in eukaryotes Base editors create precise genomic edits by directing nucleobase deamination or removal without inducing double-stranded DNA breaks. However, a vast chemical space of other DNA modifications remains to be explored for genome editing. Here we harness the bacterial antiphage toxin DarT2 to append ADP-ribosyl moieties to DNA, unlocking distinct editing outcomes in bacteria versus eukaryotes. Fusing an attenuated DarT2 to a Cas9 nickase, we program site-specific ADP-ribosylation of thymines within a target DNA sequence. In tested bacteria, targeting drives homologous recombination, offering flexible and scar-free genome editing without base replacement or counterselection. In tested yeast, plant and human cells, targeting drives substitution of the modified thymine to adenine or a mixture of adenine and cytosine with limited insertions or deletions, offering edits inaccessible to current base editors. Altogether, our approach, called append editing, leverages the addition of chemical moieties to DNA to expand current modalities for precision gene editing. Genetic engineering Synthetic biology
N Nature Biotechnology · Aug 28, 2025 Skin metatranscriptomics reveals a landscape of variation in microbial activity and gene expression across the human body Metatranscriptomics methods for the skin are hampered by low microbial biomass, contamination with host cells and low RNA stability. In this study, we developed a robust, clinically tractable skin metatranscriptomics workflow that provides high technical reproducibility of profiles, uniform coverage across gene bodies and strong enrichment of microbial mRNAs. Paired application of this protocol with metagenomics to five skin sites in a cohort of 27 healthy adults identifies a notable divergence between transcriptomic and genomic abundances. Specifically,Staphylococcusspecies and the fungiMalasseziahad an outsized contribution to metatranscriptomes at most sites, despite their modest representation in metagenomes. Species-level analysis shows signatures of microbial adaptation to their niches. Gene-level analysis identifies diverse antimicrobial genes transcribed by skin commensals in situ, including several uncharacterized bacteriocins. Correlation of microbial gene expression with organismal abundances uncovers more than 20 genes that putatively mediate interactions between microbes. This work highlights how skin metatranscriptomics identifies active species and microbial functions in situ. Biochemical reaction networks Microbiome Transcriptomics
N Nature Biotechnology · Aug 15, 2025 Scalable secondary metabolite production inStreptomycesusing a plug-and-play system Streptomycesspecies are producers of bioactive secondary metabolites with clinical, agricultural and biotechnological applications. To scale these strains for industrial production, a plug-and-play system that orchestrates multitarget engineering for maximizing cellular contributions to secondary metabolites is needed. Building on the discovery that distinct quorum-sensing receptors within theStreptomycesgenus can recognize an identical DNA-binding site, we build a quorum-sensing-triggered promoter applicable across variousStreptomyces. Integrating this promoter with a stabilizer and a multiplexer module, we develop aStreptomycesmultiplexed artificial control system (SMARTS) that converts the transient signals of diverse quorum sensing into stable and multiplexed on or off outputs with varying strengths. We build a redesigned nativeStreptomyces avermitilisfor specialized production of the nematicide baiweimectin and a de novo programmedStreptomyces venezuelaefor heterologous production of the semisynthesized antitumor drug epidoxorubicin. Notably, the baiweimectin-producing strain was scaled up to a 120-m3industrial-scale fermentation with a titer of 8.4 g L−1, underscoring the robustness of the SMARTS-based program. Applied microbiology Metabolic engineering Synthetic biology
