Overlapping genes allow multiple proteins to be encoded from a single DNA sequence, including convergent (antisense; tail-to-tail) orientations across three reading frames (phases 0, 1, and 2), with phase 1 most frequently observed in nature. Designing such overlaps is challenging due to codon degeneracy, phase-specific biases, and the need to preserve structural integrity for both proteins. Here, a purpose-built transformer encoder is introduced, trained on a balanced synthetic dataset of convergent overlaps spanning diverse prokaryotic genomes and GC contents. Controlled amino acid substitutions were incorporated during training to enhance model generalization, particularly for phase 1 overlaps. At inference, Monte Carlo dropout enabled uncertainty-aware sampling of synonymous codon solutions, which were iteratively refined using a windowed, multi-objective optimization framework. Candidate overlaps were scored using composite weighting across secondary structure preservation, substitution similarity, alignment identity, and ESM-2 contact map similarity, with SSIM applied as a rapid proxy for structural fidelity. This approach generated convergent overlaps across all phases, with phase 1 showing the highest success rates. Optimization trajectories revealed distinct dynamics, with secondary structure preservation steadily increasing despite its lower weight. External validation using SwissProt proteins stratified by AlphaFold2 pLDDT confidence supported generalization to proteins with differing rigidity, yielding high secondary structure preservation in silico. These results demonstrate that transformer models trained directly at the nucleotide level, when coupled with uncertainty-aware inference and lightweight structural proxies, can support the computational design of synthetic overlapping genes without requiring full structural prediction. This framework offers a scalable path for phase-specific, codon-aware overlap design under realistic constraints.

Bacteria have been generally greatly overlooked in the aspect of intra- and extra-cellular homeostasis, and yet, since they have evolved intricate processes and mechanisms allowing them not only to stay alive but also thrive in favorable and unfavorable environments alike, they should be considered as a close-to-ideal example of single-cell homeostasis. The bacterial responses aimed at maintaining homeostasis, while adjusting and reacting smoothly and swiftly to any changes inside and outside the cell, involve complex transcriptional networks regulated by second messengers and DNA topology, but also influenced by the presence of prophages and toxin-antitoxin systems. Their adjustment to nutrient availability also involves homeostasis in energy-related processes, such as central carbon metabolism, and crucial ion acquisition, e.g., iron. The genome stability, which is indispensable to maintain a given organisms’ functions, is achieved by control of DNA replication and repair. Furthermore, bacteria can form multicellular structures (biofilms), where homeostasis is achieved at several different levels and provides bacteria with higher chances of survival and colonization of new niches and locations. Precise correlation between the above-mentioned cellular processes makes bacteria highly intriguing objects of studies. Homeostasis is the most important basis of their life-style flexibility, thus understanding of these processes is indispensable for both: the basic and applied sciences. For example, understanding how chromosomal architecture and DNA topology coordinate global gene expression is essential for optimizing strain engineering and synthetic biology applications. Moreover, bacterial homeostasis regulatory processes can be employed as targets for antibacterial agents and prospective therapies.

Lactoferrin (LF) is a heat-sensitive, iron-binding globular glycoprotein. Glycosylation and glycation can significantly alter its three-dimensional structure, spatial conformation, and functional properties, thereby affecting its thermal stability. This study compared the thermal stability of glycosylated LF from caprine colostrum (CLF) and mature milk (MLF) with that of glycated LF from lactosylated LF (LFL). During the heating process, both CLF and MLF exhibited heat-induced aggregation in SDS-PAGE, the holo-peak of CLF was higher than that of MLF after heat treatment, and the thermal transition temperature of CLF (95 °C) was higher than that of MLF (85 °C), suggesting glycosylation plays a role in the heat stability of LF. Compared to MLF, LFL exhibited less thermal aggregation and greater retention of secondary structure. In addition, the LFL showed more rod-shaped proteins in SEM, indicating that the LFL had improved thermal stability. This study reveals the potential effects of glycosylation and glycation in enhancing the thermal stability of LF.

Ribonuclease P (RNase P), a ribozyme conserved across all domains of life, is involved in the tRNA 5′ maturation. RNase P recognizes precursor tRNA based on its structure, not the tRNA sequence. This feature had been exploited to engineer RNase P to selectively target and cleave any RNA as a gene inactivation strategy. Of these, a strategy called M1GS involves tethering an Escherichia coli M1 RNA to a short stretch of guide sequence (GS), which is complementary to the RNA targeted for cleavage. Despite its simplicity and versatility, M1GS tool appears to be underutilized compared to other gene inactivation strategies. Perhaps one of the reasons is that employment of the M1GS strategy requires prior knowledge about the requirements of the M1GS target sites. To facilitate its broader use, we have developed a Python script-based user-friendly bioinformatic tool built based on the requirements of M1GS to predict its target sites for any given RNA (using either DNA or RNA sequence as an input). In this study, we first demonstrate the utility of the bioinformatic tool in predicting M1GS target sites for human 28S rRNA and then we show that the customized M1GS-mediated downregulation of 28S rRNA in a human cancer cell line. We further validate the bioinformatic tool by predicting M1GS target sites for two previously targeted RNAs. Lastly, we discuss the utility of M1GS ribozyme-mediated rRNA downregulation as a potential anticancer modality in cancers where rRNAs are upregulated.

Polyethylene (PE) is one of the widely utilized plastics globally, valued for its durability but unsustainable due to its resistance to biodegradation in a natural environment, leading to severe environmental accumulation. Recent studies have identified microorganisms, insects, and potential PE-degrading enzymes (PEases) capable of breaking down PE, suggesting a possible route for biorecycling. However, research in this area remains in its early stages, with limited understanding of the enzymatic mechanisms involved and the degradation products formed. A major barrier lies in the chemically inert nature of PE’s carbon–carbon and carbon–hydrogen bonds, which makes enzymatic degradation particularly challenging and unlikely to occur through a single enzyme. Overcoming these limitations requires the discovery and engineering of complex enzymatic pathways, supported by emerging tools such as omics technologies, structure-guided design, and computer-aided enzyme engineering. In parallel, the biotechnological upcycling of PE waste into value-added chemicals, by first breaking down PE into smaller products and then using them as microbial feedstocks, holds significant potential but is currently underexplored. To date, polyhydroxyalkanoate (PHA) remains the most studied PE waste upcycled biopolymer product, with only a few other studies showing production of diacids, protein, wax esters, and lipids. This highlights the need for expanded research into microbial metabolism and metabolic engineering to enable more diverse and efficient PE waste bioconversion routes. This review summarizes the current state as an integrated effort for biorecycling of PE, including PE pretreatment technologies, enzymatic PE degradation, microbial PE degradation, and PE upcycling into value-added chemicals via metabolic engineering. This review also highlights key scientific challenges and outlines future directions for PE degradation and transforming PE waste into valuable and sustainable products.

Genome editing has emerged as a powerful tool for crop breeding. However, most commonly used approaches rely on tissue culture and the use of transgenic materials, which are time-consuming and labor-intensive, and often exhibit strong genotype dependency. Here we utilized tobacco rattle virus (TRV)-mediated ISYmu1 genome editing, coupled with in planta shoot regeneration, to achieve somatic and heritable genome editing in different tomato cultivars. By targeting the SlPDS gene, we successfully generated virus- and transgene-free homozygous mutant progeny in one generation. We further verified the method by generating somatic edits in SlDA1, a gene with potential agronomic relevance. Given the wide host range of TRV and its compatibility with in planta shoot regeneration, our system could be broadly applicable for rapid, non-transgenic and heritable genome editing, thereby advancing functional genomics and crop improvement.

Foodborne pathogen contamination is a major global public health issue, highlighting the need for rapid, sensitive, and portable detection technologies. Conventional methods such as microbial culture, polymerase chain reaction, and enzyme-linked immunosorbent assay are limited by complexity, low sensitivity, and high cost, restricting on-site use. Functional nucleic acids (FNAs), including aptamers and DNAzymes, have emerged as versatile molecular recognition elements for next-generation biosensors. Their programmability, stability, and specificity enable reliable real-time food monitoring. Moreover, nucleic acid-based signal amplification methods greatly enhance detection sensitivity. This review summarizes recent advances in FNA-based biosensors for foodborne pathogen detection, covering screening strategies like systematic evolution of ligands by exponential enrichment, sensing mechanisms with various readouts (fluorescence, colorimetry, electrochemistry), and innovative amplification strategies for improving practical application in food safety monitoring. Representative biosensor designs, mechanisms, and performances are highlighted. Finally, existing challenges and future perspectives for enhancing detection accuracy are discussed. FNA-driven biosensors hold great promise for preventing foodborne diseases and advancing portable, high-sensitivity detection platforms.

Deep learning has revolutionized biomolecular modeling, enabling the prediction of diverse structures with atomic accuracy. However, leveraging the atomic-level precision of the structure prediction model for de novo design remains challenging. Here, we present HalluDesign, a general all-atom framework for protein optimization and de novo design, which iteratively update protein structure and sequence. HalluDesign harnesses the inherent hallucination capabilities of AlphaFold3-style structure prediction models and enables fine-tune free, forward-pass only sequence-structure co-optimization. Structure conditioning at different noise level in the structure prediction stage allows precise control over the sampling space, facilitating tasks from local and global protein optimization to de novo design. We demonstrate the versatility of this approach by optimizing suboptimal structures, rescuing previously unsuccessful designs, designing new biomolecular interactions and generating new protein structures from scratch. Experimental characterization of binder design spanning small molecule, metal ion, protein, and antibody design of phosphorylation-specific peptide revealed high design success rates and excellent structural accuracy. Together, our comprehensive computational and experimental results highlight the broad utility of this framework. We anticipate that HalluDesign will further unlock the modeling and design potential of AlphaFold3-like models, enabling the systematic creation of complex proteins for a wide range of biotechnological applications.

Gene editing, especially the CRISPR/Cas9 system, has revolutionized trait development in crops. However, large parts of the world are missing out on applying CRISPR in planta. There is an obvious lack of gene editing applications in locally relevant crops in the Global South which tend to be neglected by mainstream agricultural research and development. Access barriers to these new breeding technologies need to be removed to allow the potential impact of these technologies on food security to happen. Here, we present the ENABLE® Gene Editing in planta toolkit, a minimal molecular toolbox allowing users to create a CRISPR knockout vector for transient or stable plant transformation in two simple cloning steps. We validate the toolkit in rice (Oryza sativa) protoplasts and in Arabidopsis thaliana plants. The ENABLE® kit is designed to be utilized specifically by users in the Global South who are new to CRISPR technology by providing a simple workflow, extensive accompanying protocols as well as options for low cost methods for cloning verification and gene editing verification in planta. We hope that our toolkit helps bridge the gap between the recent biotechnological advancements in plant breeding that high income countries can access and the lack of those technologies in low and middle income countries.

PCR enables rapid, cost-effective diagnostics but requires prior identification of genomic regions that allow sensitive and specific identification of target microbial groups, herein referred to as microbial signature sequences. We introduce Seqwin, an open-source framework designed to automate microbial genome signature discovery. Tens of thousands of microbial genomes are now available, limiting the application of existing manual and automated approaches for identifying signatures. Modern approaches that are capable of leveraging all available microbial genomes will ensure sensitive and accurate DNA signatures identification and enable robust pathogen detection for clinical, environmental, and public health applications. Seqwin builds weighted pan-genome minimizer graphs and uses a traversal algorithm to identify signature sequences that occur frequently in target genomes but remain rare in non-targets. Unlike earlier tools that depend on strict presence or absence of sequences, Seqwin accommodates natural sequence variation and scales to very large genome collections. When applied to genomes from C. difficile, M. tuberculosis, and S. enterica, Seqwin recovered more high-quality signatures than alternative methods with lower computational burden. Seqwin analysis of nearly 15,000 S. enterica genomes yielded over 200 candidate signatures in less than 10 minutes. Seqwin provides an open-source solution for the long-standing need for scalable microbial signature discovery and diagnostic assay design.  https://github.com/treangenlab/Seqwin