Efficient genome writing in mammalian cells requires robust methods for integrating large DNA payloads. The previously described method mammalian Switching Antibiotic resistance markers Progressively for Integration (mSwAP-In) enables iterative, biallelic genome rewriting in mammalian stem cells with DNA payloads exceeding 100 kb. However, the lack of standardized vectors and certain technical constraints have limited its broader adoption. Here we present an improved plasmid toolkit designed to streamline the implementation of mSwAP-In. The toolkit includes two core vectors. pLP-TK (pCTC174) is a landing-pad plasmid compatible with Golden Gate assembly of genomic homology arms and supports both mSwAP-In and the recombinase-mediated cassette exchange method Big-IN. mSwAP-In MC2v2 (pKBA135) is a versatile Big DNA assembly and delivery vector that supports Gibson-based assembly and incorporates positive, negative, and fluorescent selection markers, as well as a backbone counterselection cassette to minimize unwanted plasmid integration. The vector architecture also enables propagation in yeast and bacterial hosts, inducible plasmid copy-number amplification in standard E. coli strains, and CRISPR/Cas9-mediated payload release through preinstalled guide RNA target sites. We further characterize the FCU1/5-FC counterselection system in mouse embryonic stem cells and define conditions that minimize its bystander toxicity. Finally, we provide a set of Cas9-gRNA expression plasmids optimized for common mSwAP-In applications. Together, these reagents constitute a standardized and experimentally validated toolkit that simplifies large-scale genome writing using mSwAP-In.

Biomolecular interactions lie at the core of cellular life, spanning diverse molecular modalities from small molecules to nucleic acids and proteins. Nevertheless, design strategies remain separated despite shared physicochemical principles of molecular recognition. Here we present AnewOmni, a unified generative framework trained on more than 5 million biomolecular complexes, that enables transferable molecular design across molecular scales by assembling chemically meaningful building blocks at atomic resolution. We further introduce programmable graph prompts to support user-defined chemical, topological, and geometric steering during generation, exploring hybrid and unconventional chemistries beyond canonical structures. We demonstrate that transferable learning of interaction patterns and physical constraints across molecular modalities is possible, via an atom-to-block latent space capturing both atomic details and structural priors. The framework successfully designed small molecules, peptides, and nanobodies targeting the challenging KRAS G12D switch II pocket, as well as orthosteric peptides and allosteric small-molecule inhibitors for PCSK9 in the absence of known binding site, achieving 23%-75% success with only low-throughput validation, bypassing modality-specific high-throughput screening. AnewOmni is the first to succeed in functional molecular design across all scales, from small chemical entities to large biologics, and represents a stepstone towards general molecular reasoning engines, advocating a generative foundation model for biomolecular interactions to enter regimes where data and human intuition remain limited.

We present a living, synthetic bacterial cell made by transplanting a complete genome into a dead cell. After killing Mycoplasma capricolum cells by chemically crosslinking their genome with Mitomycin C (MMC), we installed synthetic Mycoplasma mycoides genomes into the resulting dead cells using Whole Genome Transplantation (WGT). During WGT, a synthetic donor genome is placed into a recipient cell, thereby reprogramming that cell to adopt a new genetic identity. WGT has only been demonstrated using species within one phylogenetic clade of Mollicutes bacteria. A major barrier to expanding WGT to diverse bacterial species has been the inability to inactivate the recipient genome, leading to false positive transplants due to homologous recombination of antibiotic resistance markers from the donor genome into the recipient cell genome. Here, we address this key limitation by removing reliance on an antibiotic resistance marker to select for transplants; recipient cells are dead unless revived by the installation of a new genome. Our work demonstrates a general approach to fully inactivate the recipient cell genome, reports the first living synthetic bacterial cell constructed from non-living parts, and advances WGT for building engineered or synthetic cells for diverse applications.

The acid activity of enzymes, characterized by the minimum pH at which enzymes remain active (pHmin), is crucial for industrial applications in acidic environments. However, the rational design of acid-active enzymes remains challenging due to limited understanding of sequence-structure-activity relationships under acidic conditions. Here, we propose ACENet, a graph neural network that predicts enzyme pHmin by integrating surface features of protein structures with evolutionary representations derived from the large-scale protein language model ESM-2. ACENet achieved a Pearson correlation coefficient of 0.85 on the test dataset, significantly outperforming other deep learning baseline models and maintains stable pHmin predictions under various conditions. Even on a subset of the dataset with less than 20% homology, the PCC remains above 0.5, with an RMSE (Root mean square error) less than 1.4. ACENet also present excellent performance in the annotation of pHmin for homologous proteins and the predictive screening of minimal active pH in protein mutants. Remarkably, ACENet could identify the catalytic region as key determinants of acid activity through residue-level interpretability analysis. Overall, ACENet accelerates the development of highly efficient biocatalysts for diverse applications where acidic conditions predominate.

Delivery of biomolecules into plant vascular tissues remains a barrier to managing diseases caused by insect vector-borne pathogens and to modifying phenotypes of established perennial crops. Inspired by the vascularized growth of crown galls induced by Agrobacterium tumefaciens, we repurposed the bacterium’s plant growth regulator (PGR) genes to engineer autonomously dividing, transgene-expressing plant cell structures termed symbionts. A plant transformation vector (pSYM) incorporating the IaaM, IaaH, Ipt and gene5 cassette from A. tumefaciens strain C58 together with a gene of interest on the same transfer DNA was delivered to stems of herbaceous and woody dicots using disarmed A. tumefaciens strain EHA105. Symbiont morphology, vascular differentiation, transgene expression, molecular mobility and protein secretion were evaluated using microscopy, fluorescent reporters, dye tracing, RNA silencing assays and mass spectrometry-based proteomics. pSym inoculation reproducibly generated symbionts across diverse host plant species that were vascularly integrated into their host plants and transgene expression ranging from heterogeneous niches to more uniform patterns. Small molecules moved between symbionts and host vascular tissues, whereas larger proteins exhibited more restricted mobility. Post-transcriptional gene silencing signals moved freely throughout the symbiont and slightly into adjacent stem tissue. Under tested field and greenhouse conditions in potato and tomato, respectively, gall or symbiont formation had no negative impacts on plant growth or tuber and fruit yield. In vitro, symbiont cultures abundantly secreted recombinant protein into surrounding media. Together, these results establish symbionts as a modular, plant bioengineering platform capable of producing and potentially delivering biomolecules without modifying the host plant genome, providing a foundation for vascular-targeted therapeutics and phenotype modulation in crops.

Protein complexes are central to cellular function, but experimental determination of their structures remains challenging. Structure prediction methods require prior knowledge of stoichiometry - the number of copies of each protein entity within a complex. Current approaches rely on computationally expensive brute-force methods that run structure prediction on multiple stoichiometry combinations, often with limited accuracy.  We introduce Stoic, a method that uses protein language model embeddings to predict protein complex stoichiometry. Our approach learns to identify interface residues that participate in protein-protein interactions, rather than relying on global sequence features. By integrating these interface-aware embeddings into a graph neural network, Stoic achieves fast and accurate stoichiometry prediction for both homomeric and heteromeric targets. Availability: Source code for inference and training along with web versions are available in the repository at https://github.com/PickyBinders/stoic.

Pseudomonas chlororaphis ATCC 9446 is a non-pathogenic rhizobacterium with biotechnological relevance as a biocontrol agent and a promising chassis for synthetic biology. Understanding which genes are strictly required for survival is fundamental to both bacterial physiology and chassis engineering. Here, we generate a genome-wide map of genetic essentiality for P. chlororaphis using high-density Random Barcoded Transposon Mutagenesis (RB-TnSeq). To convert gene-level annotation into biological insight, we layered functional assignments from complementary annotation pipelines, integrating orthology/domain classifiers, ontology mapping, protein export, and delineation of secondary metabolism. These overlays reveal that essentiality concentrates in canonical information processing, envelope biogenesis, and central energy/cofactor and nucleotide metabolism, while large genomic regions are non-essential and therefore represent safe candidates for streamlining and pathway installation. Mapping essentiality onto biosynthetic gene clusters (BGCs) shows that most pathways are dispensable, but some essential genes co-localize, clarifying boundaries for safe editing around BGC loci. Comparison of experimentally determined essential genes with in silico predictions across additional P. chlororaphis genomes show strong overall agreement. Conversely, 32 essential gene orthogroups were found to be conserved across most genomes, yet were classified as non-essential by a prediction algorithm. Together, the resolved essential genome and its integrative functional interpretation provide a durable reference for P. chlororaphis biology and a functional blueprint that can be leveraged for rational streamlining in agricultural, biocontrol and industrial biotechnology.

Bacteriophages (phages) are being cataloged at unprecedented rates and are recognized as key players in nutrient and energy cycling across ecosystems. Yet, the biological determinants of phage-host specificity and infection success remain poorly understood. Here, we used a randomly barcoded, genome-wide, loss-of-function transposon mutant library (RB-TnSeq) of Klebsiella sp. M5al, a plant-associated, nitrogen-fixing rhizobacterium, to identify bacterial genetic determinants necessary for the infection of 25 double-stranded DNA phages spanning five viral families. This approach identified 42 bacterial genes associated with phage infection, encompassing genes involved in receptor biosynthesis, regulatory pathways, electron transport, and unknown functions. Disruption of genes involved in receptor biosynthesis, such as glycosyltransferases involved in lipopolysaccharide (LPS) production, conferred cross-resistance to half of the phages, while intracellular gene disruptions had phage-specific effects. Taxonomically, we found significant differences in RB-TnSeq profiles across phage families and genera. While bacterial gene requirements are generally clustered by phage genus, we also observed notable divergences within genera. These differences were associated with variation in receptor usage, likely driven by divergence in tail fiber or other host recognition proteins. In some cases, differential reliance on bacterial intracellular factors suggested that even closely related phages may depend on distinct infection strategies, potentially involving phage-specific genes of unknown function. Our findings reveal key genetic dependencies shaping phage-host interactions and offer a framework for predicting infection outcomes in natural and applied settings.