bioinnovationbeacon.com

Author: admin

  • Engineering smart delivery for gene editors

    Engineering smart delivery for gene editors

    Modern genome editing techniques, including CRISPR systems, hold great potential for treating genetic diseases. However, delivering these molecular tools reliably to their target cells remains a significant challenge.

    “Previous viral and non-viral delivery systems such as adeno-associated viruses (AAVs), lipid nanoparticles (LNPs), and other virus-like particles (VLPs), have been valuable but face limitations,” says Dr. Dong-Jiunn Jeffery Truong, last author of the study and group leader at the Institute for Synthetic Biomedicine at Helmholtz Munich. “Challenges include the increased persistence of gene editors potentially causing immune reactions, or simply their limited efficiency. ENVLPE directly addresses these issues while its modular design maintains compatibility with future gene-editing advancements.”

    ENVLPE is based on modified, non-infectious virus-derived shells. These act as carriers for molecular gene editors such as base or prime editors — specialized CRISPR tools that can chemically change single DNA bases in the genome and remove or insert new DNA sequences. ENVLPE’s design solves the logistics challenge of previous methods during the production of the VLPs by hijacking the intracellular transport mechanism so that all components come together at the right time and place.

    Prior methods often included partially assembled, non-functional gene editors, reducing delivery effectiveness. “ENVLPE now not only ensures the packaging of fully assembled gene editors but also contains an extra molecular shield that protects the most vulnerable part of the editor from degradation during transport,” explains Truong. “This allows the genetic tools to be safely delivered into target cells where the intended DNA edit can take place.”

    Restoring Vision: Gene Editing in Action

    In close collaboration with a team led by Prof. Krzysztof Palczewski, a professor of ophthalmology at UC Irvine, the scientists tested the ENVLPE system in a mouse model of inherited blindness. “The mice carry a disabling mutation in the Rpe65 gene, which is essential for producing light-sensitive molecules in the retina, and therefore are fully blind and unresponsive to light,” explains Samuel W. Du, a co-author and MD/PhD candidate at UC Irvine. After injecting ENVLPE into the subretinal space (the area between the retinal pigment epithelium and photoreceptors) to correct the mutation, the animals began to respond to light stimuli again. “The extent of restoration was astounding,” says Julian Geilenkeuser, co-first author of the study and a doctoral researcher at the Institute for Synthetic Biomedicine. “It showed us that our particles have real therapeutic potential in a living animal.”

    Compared to established systems, ENVLPE achieved significantly better results: a competing system required more than 10 times the dose to reach similar effects. “Our goal was to build a tool that is both useful for researchers and suitable for real-world applications,” says Niklas Armbrust, also co-first author and a doctoral researcher at the Institute for Synthetic Biomedicine. “We resolved critical bottlenecks and achieved a much more efficient packaging by the delivery agents.”

    Advancing Cancer Therapy with Universal T Cells

    ENVLPE could also open up new possibilities for adoptive T cell therapies, where immune cells taken from the patient are genetically modified in the lab so that they can specifically recognize and attack tumor cells. In collaboration with Dr. Andrea Schmidts’ laboratory at TUM University Hospital, ENVLPE facilitated the targeted removal of specific surface molecules that could trigger an immune response when the cells are administered to a recipient different from the donor. This could lead to the development of so-called “universal” T cells that do not need to be customized for individual patients, making treatments more accessible and cost-effective.

    These innovations address critical challenges in both in vivo gene therapies for genetically inherited diseases and ex vivo cell therapies for cancer, paving the way for important translational advancements. “The highly modular ENVLPE system brings us substantially closer to on-demand and precise genetic modifications of complex cellular models,” says Prof. Gil Westmeyer, Director of the Institute for Synthetic Biomedicine and Professor for Neurobiological Engineering at TUM and co-senior author of the study. “It is an example of how synthetic biology can help drive medical innovation.”

    Moving Toward Clinical Use

    Having now achieved highly efficient delivery of the most common gene-editing tools, the team now seeks to use the diversity found in nature, along with the recent advancements in AI-assisted protein design, to increase targeting precision by restricting the delivery of these tools to specific cell or tissue types only. To move ENVLPE toward clinical application, the research team is pursuing follow-up funding from translational grants and partnerships in the pharmaceutical industry. The goal is to optimize the technology for various therapeutic applications and ultimately make it available to patients.

    What is ENVLPE?

    ENVLPE stands for “Engineered Nucleocytosolic Vehicles for Loading of Programmable Editors.” These are non-infectious virus-like particles that can efficiently transport gene-editing tools such as base or prime editors into target cells. ENVLPE overcomes two major limitations of earlier systems: the instability of the guide RNA payload and inefficient packaging of only functional gene editors within the production cells. By adjusting intracellular transport mechanisms, the researchers significantly increased the system’s efficiency and safety.

  • Powerful new software platform could reshape biomedical research by making data analysis more accessible

    Powerful new software platform could reshape biomedical research by making data analysis more accessible

    A powerful new software platform called the Playbook Workflow Builder is set to transform biomedical research by allowing scientists to conduct complex and customized data analyses without advanced programming skills. An article that describes the new platform was published in the April 3online issue of the journal PLOS Computational Biology.

    Developed by a multi-institutional team that was led by Icahn School of Medicine at Mount Sinai investigators as part of the National Institutes of Health Common Fund Data Ecosystem (CFDE) program, researchers from across the United States developed the web-based platform that enables scientists to analyze and visualize their own data independently through an intuitive, interactive interface.

    Traditionally, experimental biologists rely heavily on bioinformaticians to process and analyze the large datasets they collect. The new software platform changes this paradigm by offering a modular, user-friendly system where scientists can design custom workflows using pre-built analytical components — akin to assembling “LEGO® pieces,” say the investigators.

    “This approach could fundamentally reinvent data analysis in biomedical research and beyond. By allowing researchers to conduct sophisticated analyses without requiring advanced programming skills, it removes a major barrier to data-driven discovery,” says first author Daniel J.B. Clarke, MS, a data scientist in the lab of senior corresponding author Avi Ma’ayan, PhD, Professor, Pharmacological Sciences, and Director of the Mount Sinai Center for Bioinformatics at the Icahn School of Medicine at Mount Sinai. “With greater accessibility and efficiency, scientists across disciplines can explore complex datasets more easily, uncover new insights faster, and ultimately accelerate the pace of scientific publication and discovery.”

    Users of the Playbook Workflow Builder engage with an intuitive user interface, clicking on cards to construct their data analysis workflows. Alternatively, users can interface with a large language model-powered chatbot to design and build their data analysis pipelines. As they upload their data and input their instructions, the system automatically generates detailed documentation, including interactive figures, clear figure legends, and step-by-step automatically generated method descriptions.

    This ensures that the entire workflow is well-organized and easy to follow. Additionally, researchers can export their workflows in multiple formats, making it easier to share their work with colleagues, which also enhances reproducibility across different studies.

    Currently, the team is actively working to expand the platform’s capabilities by adding more analytical components, refining the user experience, and fostering a larger community of users and contributors.

    “We see this as just the beginning. By reducing reliance on bioinformatics specialists, this platform has the potential to dramatically accelerate the pace of research by enabling experimental biologists to perform sophisticated analyses independently,” says Dr. Ma’ayan “As more scientists adopt this tool, we expect it to streamline workflows, speed up the sharing of findings, and inspire new AI-driven systems that further enhance scientific reasoning and data interpretation. In the years to come, this could become an essential resource for biomedical research, unlocking new insights and driving innovation across the field.”

  • Microbial cell factories for sustainable chemical production

    Microbial cell factories for sustainable chemical production

    Climate change and the depletion of fossil fuels have raised the global need for sustainable chemical production. In response to these environmental challenges, microbial cell factories are gaining attention as eco-friendly platforms for producing chemicals using renewable resources, while metabolic engineering technologies to enhance these cell factories are becoming crucial tools for maximizing production efficiency. However, difficulties in selecting suitable microbial strains and optimizing complex metabolic pathways continue to pose significant obstacles to practical industrial applications.

    KAIST (President Kwang-Hyung Lee) announced on 27th of March that Distinguished Professor Sang Yup Lee’s research team in the Department of Chemical and Biomolecular Engineering comprehensively evaluated the production capabilities of various industrial microbial cell factories using in silico simulations and, based on these findings, identified the most suitable microbial strains for producing specific chemicals as well as optimal metabolic engineering strategies.

    Previously, researchers attempted to determine the best strains and efficient metabolic engineering strategies among numerous microbial candidates through extensive biological experiments and meticulous verification processes. However, this approach required substantial time and costs. Recently, the introduction of genome-scale metabolic models (GEMs), which reconstruct the metabolic networks within an organism based on its entire genome information, has enabled systematic analysis of metabolic fluxes via computer simulations. This development offers a new way to overcome limitations of conventional experimental approaches, revolutionizing both strain selection and metabolic pathway design.

    Accordingly, Professor Lee’s team at the Department of Chemical and Biomolecular Engineering, KAIST, evaluated the production capabilities of five representative industrial microorganisms — Escherichia coli, Saccharomyces cerevisiae, Bacillus subtilis, Corynebacterium glutamicum, and Pseudomonas putida — for 235 bio-based chemicals. Using GEMs, the researchers calculated both the maximum theoretical yields and the maximum achievable yields under industrial conditions for each chemical, thereby establishing criteria to identify the most suitable strains for each target compound.

    The team specifically proposed strategies such as introducing heterologous enzyme reactions derived from other organisms and exchanging cofactors used by microbes to expand metabolic pathways. These strategies were shown to increase yields beyond the innate metabolic capacities of the microorganisms, resulting in higher production of industrially important chemicals such as mevalonic acid, propanol, fatty acids, and isoprenoids.

    Moreover, by applying a computational approach to analyze metabolic fluxes in silico, the researchers suggested strategies for improving microbial strains to maximize the production of various chemicals. They quantitatively identified the relationships between specific enzyme reactions and target chemical production, as well as the relationships between enzymes and metabolites, determining which enzyme reactions should be up- or down-regulated. Through this, the team presented strategies not only to achieve high theoretical yields but also to maximize actual production capacities.

    Dr. Gi Bae Kim, the first author of this paper from the KAIST BioProcess Engineering Research Center, explained, “By introducing metabolic pathways derived from other organisms and exchanging cofactors, it is possible to design new microbial cell factories that surpass existing limitations. The strategies presented in this study will play a pivotal role in making microbial-based production processes more economical and efficient.” In addition, Distinguished Professor Sang Yup Lee noted, “This research serves as a key resource in the field of systems metabolic engineering, reducing difficulties in strain selection and pathway design, and enabling more efficient development of microbial cell factories. We expect it to greatly contribute to the future development of technologies for producing various eco-friendly chemicals, such as biofuels, bioplastics, and functional food materials.”

    This research was conducted with the support from the Development of platform technologies of microbial cell factories for the next-generation biorefineries project and Development of advanced synthetic biology source technologies for leading the biomanufacturing industry project (Project Leader: Distinguished Professor Sang Yup Lee, KAIST) from National Research Foundation supported by the Korean Ministry of Science and ICT.

  • how bacteria ‘vaccinate’ themselves with genetic material from dormant viruses

    how bacteria ‘vaccinate’ themselves with genetic material from dormant viruses

    Like people, bacteria get invaded by viruses. In bacteria, the viral invaders are called bacteriophages, derived from the Greek word for bacteria-eaters, or in shortened form, “phages.” Scientists have sought to learn how the single-cell organisms survive phage infection in a bid to further understand human immunity and develop ways to combat diseases.

    Now, Johns Hopkins Medicine scientists say they have shed new light on how bacteria protect themselves from certain phage invaders — by seizing genetic material from weakened, dormant phages and using it to “vaccinate” themselves to elicit an immune response.

    In their experiments, the scientists say Streptococcus pyogenes bacteria (which cause strep throat) take advantage of a class of phages known as temperate phages, which can either kill cells or become dormant. The bacteria steal genetic material from temperate phages during this dormant period and form a biological “memory” of the invader that their offspring inherit as the bacteria multiply. Equipped with these memories, the new population can recognize these viruses and fight them off.

    A report on the experiments, supported in part by the National Institutes of Health, was published March 12 in the journal Cell Host & Microbe. The findings help scientists better understand how bacterial cells that cause serious diseases, including Staph and E. coli infections and cholera, become toxic to humans — a process that involves toxic genes expressed by otherwise dormant phages that reside within the bacterial cell, says corresponding author Joshua Modell, Ph.D., associate professor of molecular biology and genetics at the Johns Hopkins University School of Medicine.

    “We essentially wanted to answer the question: If bacterial cells don’t have any memory, or survival skills, to combat a new temperate phage that shows up, how do they buy themselves enough time to establish a new memory, before they succumb to that initial infection?” says Modell.

    The Johns Hopkins investigators say bacteria have long been known to use CRISPR-Cas systems to chop up phage DNA, break it down and get rid of it. Crucially, CRISPR systems can only destroy DNA that matches a “memory” captured from a prior infection and stored within the bacteria’s own genome, say the researchers. In this way, the CRISPR system acts as a recording device that documents the long list of foreign invaders encountered by a particular bacterial strain.

    To conduct their research, the scientists say they infected populations of bacteria with naturally occurring phages that go dormant or genetically engineered non-dormant phages in separate flasks that contained millions of bacterial cells.

    “Our results indicate that the bacteria’s CRISPR system was more effective at using the naturally dormant phage to pull parts of the viral genetic code into their genome,” says Modell. “When we tested phages that could not go dormant, the CRISPR system did not work nearly as well.”

    After isolating the bacteria that survived, and letting the survivors repopulate the flask, the scientists used genome sequencing to catalog hundreds of thousands of new DNA memories that the CRISPR Cas9 system had created from the test phages, honing in on those that contribute to cell immunity. The scientists also determined that bacteria created those memories during the temperate phage’s dormancy period, when it did not pose a threat to the population.

    “This is conceptually similar to a vaccine with an attenuated virus,” says Nicholas Keith, a graduate student and first author of the paper. “We believe this is the reason why the CRISPR Cas9 system has a unique relationship with this specific class of temperate phage.”

    “We can use these types of experiments to find what elements of the phage, the bacterial host and its CRISPR system are important for all stages of bacterial immunity,” Keith says.

    In future experiments, the scientists aim to learn more about how CRISPR systems protect bacteria cells from viruses that don’t go dormant, Modell says.

    “We know CRISPR systems are one of the first lines of defense against the transfer of hazardous genes from phages that turn bacterial cells toxic,” says Modell. “Furthermore, our studies will inform the design of ‘phage therapies’ which could be used in clinical cases where a bacterial infection is resistant to all available antibiotics.”

    In addition to Keith and Modell, study contributors are Rhett Snyder from Johns Hopkins and Chad Euler from Hunter College.

    The research was funded by the Johns Hopkins University School of Medicine, the National Institutes of Health National Institute of General Medical Sciences (R35GM142731), the Rita Allen Foundation and the National Science Foundation.

  • New AI tool generates high-quality images faster than state-of-the-art approaches

    New AI tool generates high-quality images faster than state-of-the-art approaches

    The ability to generate high-quality images quickly is crucial for producing realistic simulated environments that can be used to train self-driving cars to avoid unpredictable hazards, making them safer on real streets.

    But the generative AI techniques increasingly being used to produce such images have drawbacks. One popular type of model, called a diffusion model, can create stunningly realistic images but is too slow and computationally intensive for many applications. On the other hand, the autoregressive models that power LLMs like ChatGPT are much faster, but they produce poorer-quality images that are often riddled with errors.

    Researchers from MIT and NVIDIA developed a new approach that brings together the best of both methods. Their hybrid image-generation tool uses an autoregressive model to quickly capture the big picture and then a small diffusion model to refine the details of the image.

    Their tool, known as HART (short for Hybrid Autoregressive Transformer) can generate images that match or exceed the quality of state-of-the-art diffusion models, but do so about nine times faster.

    The generation process consumes fewer computational resources than typical diffusion models, enabling HART to run locally on a commercial laptop or smartphone. A user only needs to enter one natural language prompt into the HART interface to generate an image.

    HART could have a wide range of applications, such as helping researchers train robots to complete complex real-world tasks and aiding designers in producing striking scenes for video games.

    “If you are painting a landscape, and you just paint the entire canvas once, it might not look very good. But if you paint the big picture and then refine the image with smaller brush strokes, your painting could look a lot better. That is the basic idea with HART,” says Haotian Tang PhD ’25, co-lead author of a new paper on HART.

    He is joined by co-lead author Yecheng Wu, an undergraduate student at Tsinghua University; senior author Song Han, an associate professor in the Department of Electrical Engineering and Computer Science (EECS), a member of the MIT-IBM Watson AI Lab, and a distinguished scientist of NVIDIA; as well as others at MIT, Tsinghua University, and NVIDIA. The research will be presented at the International Conference on Learning Representations.

    The best of both worlds

    Popular diffusion models, such as Stable Diffusion and DALL-E, are known to produce highly detailed images. These models generate images through an iterative process where they predict some amount of random noise on each pixel, subtract the noise, then repeat the process of predicting and “de-noising” multiple times until they generate a new image that is completely free of noise.

    Because the diffusion model de-noises all pixels in an image at each step, and there may be 30 or more steps, the process is slow and computationally expensive. But because the model has multiple chances to correct details it got wrong, the images are high-quality.

    Autoregressive models, commonly used for predicting text, can generate images by predicting patches of an image sequentially, a few pixels at a time. They can’t go back and correct their mistakes, but the sequential prediction process is much faster than diffusion.

    These models use representations known as tokens to make predictions. An autoregressive model utilizes an autoencoder to compress raw image pixels into discrete tokens as well as reconstruct the image from predicted tokens. While this boosts the model’s speed, the information loss that occurs during compression causes errors when the model generates a new image.

    With HART, the researchers developed a hybrid approach that uses an autoregressive model to predict compressed, discrete image tokens, then a small diffusion model to predict residual tokens. Residual tokens compensate for the model’s information loss by capturing details left out by discrete tokens.

    “We can achieve a huge boost in terms of reconstruction quality. Our residual tokens learn high-frequency details, like edges of an object, or a person’s hair, eyes, or mouth. These are places where discrete tokens can make mistakes,” says Tang.

    Because the diffusion model only predicts the remaining details after the autoregressive model has done its job, it can accomplish the task in eight steps, instead of the usual 30 or more a standard diffusion model requires to generate an entire image. This minimal overhead of the additional diffusion model allows HART to retain the speed advantage of the autoregressive model while significantly enhancing its ability to generate intricate image details.

    “The diffusion model has an easier job to do, which leads to more efficiency,” he adds.

    Outperforming larger models

    During the development of HART, the researchers encountered challenges in effectively integrating the diffusion model to enhance the autoregressive model. They found that incorporating the diffusion model in the early stages of the autoregressive process resulted in an accumulation of errors. Instead, their final design of applying the diffusion model to predict only residual tokens as the final step significantly improved generation quality.

    Their method, which uses a combination of an autoregressive transformer model with 700 million parameters and a lightweight diffusion model with 37 million parameters, can generate images of the same quality as those created by a diffusion model with 2 billion parameters, but it does so about nine times faster. It uses about 31 percent less computation than state-of-the-art models.

    Moreover, because HART uses an autoregressive model to do the bulk of the work — the same type of model that powers LLMs — it is more compatible for integration with the new class of unified vision-language generative models. In the future, one could interact with a unified vision-language generative model, perhaps by asking it to show the intermediate steps required to assemble a piece of furniture.

    “LLMs are a good interface for all sorts of models, like multimodal models and models that can reason. This is a way to push the intelligence to a new frontier. An efficient image-generation model would unlock a lot of possibilities,” he says.

    In the future, the researchers want to go down this path and build vision-language models on top of the HART architecture. Since HART is scalable and generalizable to multiple modalities, they also want to apply it for video generation and audio prediction tasks.

    This research was funded, in part, by the MIT-IBM Watson AI Lab, the MIT and Amazon Science Hub, the MIT AI Hardware Program, and the National Science Foundation. The GPU infrastructure for training this model was donated by NVIDIA.

  • New CRISPR tool enables more seamless gene editing — and improved disease modeling

    New CRISPR tool enables more seamless gene editing — and improved disease modeling

    Advances in the gene-editing technology known as CRISPR-Cas9 over the past 15 years have yielded important new insights into the roles that specific genes play in many diseases. But to date this technology — which allows scientists to use a “guide” RNA to modify DNA sequences and evaluate the effects — is able to target, delete, replace, or modify only single gene sequences with a single guide RNA and has limited ability to assess multiple genetic changes simultaneously.

    Now, however, Yale scientists have developed a series of sophisticated mouse models using CRISPR (“clustered regularly interspaced short palindromic repeats”) technology that allows them to simultaneously assess genetic interactions on a host of immunological responses to multiple diseases, including cancer.

    The findings were published March 20 in the journal Nature Biomedical Engineering.

    Gene editing technologies allow scientists to use enzymes — in this case, Cas9 (CRISPR-associated protein 9) — as a sort of molecular scissors that can precisely cut or modify portions of DNA or RNA, revealing insights into the role these genes play in a variety of disorders.

    The new tool, which is called CRISPR-Cas12a, can help researchers simultaneously assess the impact of multiple genetic changes involved in variety of immune system responses, the researchers say.

    “We have created four new Cas12a mouse lines that will allow researchers to study complex genetic interactions and their effects involved in many disorders,” said Sidi Chen, an associate professor of genetics and neurosurgery at Yale School of Medicine and a pioneer in the field of CRISPR technology. He is co-corresponding author of the new study.

    With these powerful tool strains, Chen’s lab was able to induce and track changes in a variety of immune system cells in response to gene editing, and to fine-tune sets of genes in different directions simultaneously. The lab also enabled the rapid generation of new disease and treatment models, such as genetic disease in the liver, lung cancer, and skin cancer.

    This advance, he said, will offer a valuable tool to researchers creating new therapies for a host of pathologies, including cancer, metabolic disease, autoimmune disease, and neurological disorders.

    Chen is a member of the Systems Biology Institute at Yale’s West Campus and an affiliate of Yale Cancer Center, Yale Stem Cell Center, and Yale Center for Biomedical Data Science.

    Matthew B. Dong and Xiaoyu Zhou, both of Yale’s Department of Genetics, are co-corresponding authors. Kaiyuan Tang and Liqun Zhou are co-lead authors.

    The research was funded by the National Institutes of Health and the U.S. Department of Defense.

  • Universal RNA barcoding system for tracking gene transfer in bacteria

    Universal RNA barcoding system for tracking gene transfer in bacteria

    In the microscopic world of bacteria, gene transfer is a powerful mechanism that can alter cellular function, drive antibiotic resistance and even shape entire ecosystems. Now an interdisciplinary group of researchers at Rice University has developed an innovative RNA “barcoding” method to track these genetic exchanges in microbial communities, providing new insights into how genes move across species. The findings were recently published in Nature Biotechnology.

    “We’ve long known that bacteria swap genes in ways that impact human health, biotechnology and environmental stability,” said James Chappell, associate professor of biosciences and bioengineering. “But mapping which microbes participate in gene transfer has been challenging. This new technique gives us a direct way to record this information inside the cells themselves.”

    Traditional methods for studying gene transfer involve labeling mobile genetic elements with fluorescent proteins or antibiotic resistance genes. While effective, these approaches require isolating and growing microbes in a lab, limiting their use in complex environments.

    To address this challenge, an interdisciplinary team from the research labs of Rice’s Chappell, Joff Silberg and Lauren Stadler created a new synthetic biology tool. This team was composed of Matthew Dysart, Kiara Reyes Gamas, Lauren Gambill, Prashant Kalvapalle, Li Chieh Lu and August Staubus.

    The Rice team’s new method, called RNA-addressable modification (RAM), bypasses these hurdles by using a synthetic catalytic RNA (cat-RNA) to “barcode” ribosomal RNA (rRNA) inside living cells.

    By writing genetic information directly into the 16S rRNA — a molecule universally found in bacteria — the researchers could track which microbes acquired foreign DNA without disrupting their natural environment. Additionally, as targeted sequencing of 16S rRNA is the gold standard for identifying different species of bacteria, this method can leverage established and easy to use protocols and analysis software.

    “This is a game-changer for creating a mobile DNA atlas,” said Silberg, the Stewart Memorial Professor of BioSciences and professor of bioengineering. “Instead of writing information randomly in bacterial DNA, which is permanent and arduous to read out, we write information in a region of RNA that is highly conserved across the tree of life, making the information cheap and easy to read out.”

    To achieve this, the researchers designed a small ribozyme-based RNA molecule (also called catalytic RNA) that attached a unique barcode to 16S rRNA upon gene transfer. This cat-RNA was introduced into a model microbial community using conjugative plasmids, which are naturally occurring gene carriers in bacteria.

    The experiment involved introducing these barcoding plasmids into E. coli donor bacteria, which then transferred their genetic material to various microbes in a wastewater community. After 24 hours, the researchers extracted total RNA and sequenced the barcoded 16S rRNA.

    “What we saw was remarkable,” said Stadler, associate professor of civil and environmental engineering. “Around half of the bacterial taxa in the wastewater community could pick up the plasmids, giving us a detailed map of horizontal gene transfer events.”

    The study also showed that RAM can be used to measure differences in host ranges between DNA plasmid types. With tens of thousands of different DNA plasmids found in natural environmental microbes, RAM provides an easy and cost-effective method to begin to understand the relationship between plasmids and their hosts.

    “RAM can be used to track the movement of multiple genetic elements across an entire microbial community,” Chappell said. “This allowed us to track the movement of multiple plasmids in a single experiment and could be extended to study the dynamics of plasmid transfer in microbial communities and interactions between mobile genetic elements.”

    The RAM method has possible far-reaching applications in medicine, biotechnology and environmental science. One of the most pressing concerns is antibiotic resistance, as tracking how resistant genes spread in hospitals and wastewater could help predict and prevent outbreaks of drug-resistant infections. In the field of bioremediation and waste management, this technology has the potential to engineer microbiomes that efficiently break down pollutants while ensuring that beneficial genetic modifications remain contained. Additionally, in synthetic biology and biotechnology, the ability to program microbiomes for specific tasks, such as producing biofuels or pharmaceuticals, relies on safe and controlled gene transfer.

    “The potential here is enormous,” Stadler said. “We now have a way to study how bacteria share genes in their natural habitat without needing to grow them in a lab. That opens the door for a new wave of microbial research and synthetic biology applications.”

    In the future, this barcoding technique could also be expanded and applied to other forms of gene exchange such as transduction (via bacteriophages) and transformation (direct DNA uptake). Additionally, optimizing cat-RNA stability and increasing the number of unique barcodes could allow even finer resolution in tracking microbial interactions.

    “With further development, RNA barcoding could become a universal tool for storing information in environmental communities about additional microbial behaviors beyond gene transfer,” Silberg said.

  • Most current AI struggles to read clocks and calendars

    Most current AI struggles to read clocks and calendars

    Some of the world’s most advanced AI systems struggle to tell the time and work out dates on calendars, a study suggests.

    While AI models can perform complex tasks such as writing essays and generating art, they have yet to master some skills that humans carry out with ease, researchers say.

    A team from the University of Edinburgh has shown that state-of-the-art AI models are unable to reliably interpret clock-hand positions or correctly answer questions about dates on calendars.

    Unlike simply recognising shapes, understanding analogue clocks and calendars requires a combination of spatial awareness, context and basic maths – something that remains challenging for AI, the team says.

    Overcoming this could enable AI systems to power time-sensitive applications like scheduling assistants, autonomous robots and tools for people with visual impairments, researchers say.

    The team tested if AI systems that process text and images – known as multimodal large language models (MLLMs) – can answer time-related questions by looking at a picture of a clock or a calendar.

    Researchers tested various clock designs, including some with Roman numerals, with and without second hands, and different coloured dials.

    Their findings show that AI systems, at best, got clock-hand positions right less than a quarter of the time. Mistakes were more common when clocks had Roman numerals or stylised clock hands. AI systems also did not perform any better when the second hand was removed, suggesting there are deep-seated issues with hand detection and angle interpretation, the team says.

    The researchers asked AI models to answer a range of calendar-based questions, such as identifying holidays and working out past and future dates. The team found that even the best-performing AI model got date calculations wrong one-fifth of the time.

    The findings are reported in a peer-reviewed paper that will be presented at the Reasoning and Planning for Large Language Models workshop at The Thirteenth International Conference on Learning Representations (ICLR) in Singapore on 28 April 2025.

    Rohit Saxena, of the University of Edinburgh’s School of Informatics, who led the study, said: “Most people can tell the time and use calendars from an early age. Our findings highlight a significant gap in the ability of AI to carry out what are quite basic skills for people. These shortfalls must be addressed if AI systems are to be successfully integrated into time-sensitive, real-world applications, such as scheduling, automation and assistive technologies.”

    Aryo Gema, also of the School of Informatics, said: “AI research today often emphasises complex reasoning tasks, but ironically, many systems still struggle when it comes to simpler, everyday tasks. Our findings suggest it’s high time we addressed these fundamental gaps. Otherwise, integrating AI into real-world, time-sensitive applications might remain stuck at the eleventh hour.”

  • New CRISPRs expand upon the original’s abilities

    New CRISPRs expand upon the original’s abilities

    Researchers at Duke University and North Carolina State University have discovered a handful of new CRISPR-Cas systems that could add to the capabilities of the already transformational gene editing and DNA manipulation toolbox.

    Of the new recruits, one system from bacteria commonly found in dairy cows shows particular promise for human health. Its efficiency is on par with the original and most widely used CRISPR-Cas system, but its small size allows it to be more easily packaged for delivery to human cells. It also can target specific gene sequences that other systems cannot, and human immune systems are unlikely to have been exposed to it.

    The results appear online March 14 in the Proceedings of the National Academy of Sciences (PNAS).

    CRISPR-Cas9 burst onto the broader scientific scene in 2012 when a team led by Jennifer Doudna showed that it could be modified to target and cut specific segments of DNA. The CRISPR half of the system acts as a genetic homing device, while the Cas9 portion works like a scalpel that cuts where CRISPR tells it to. That work — and most of the subsequent research using CRISPR — is built on a viral defense mechanism from the bacterial species Streptococcus pyogenes. But other bacterial species also have similar defense systems with a wide range of potential abilities and limitations.

    “It’s actually remarkable that the first CRISPR-Cas systems researchers used on human cells is still the one that works the best,” said Charlie Gersbach, the John W. Strohbehn Distinguished Professor of Biomedical Engineering at Duke. “We wanted to scour bacteria found in more obscure settings for different CRISPR systems that might have different abilities.”

    To do this, Gersbach and his group turned to one of the world’s first and foremost experts on CRISPR, whose lab is just 25 miles down the highway. Seven years before the technology’s Nobel-award-winning paper, Rodolphe Barrangou of NC State characterized CRISPR as a defense system in bacteria used in dairy starter cultures. Since then, his laboratory has focused on exploring the diversity of CRISPR biology for applications ranging from food manufacturing and probiotics to manipulating wood properties by editing tree genomes.

    “There is a lot more CRISPR-Cas system diversity in nature than people appreciate, and it can be very useful to mine for diverse effectors with functional potential as molecular machines,” said Barrangou, the Todd R. Klaenhammer Distinguished Professor of Food, Bioprocessing and Nutrition Sciences at NC State. “While some incumbent effectors like Cas9 have shown great potential in the clinic already, we need to expand the CRISPR toolbox for next-generation manipulation of the genome, transcriptome and epigenome.”

    Barrangou has developed computational processes for identifying CRISPR-Cas systems within large databases of bacterial genomes. Called “CRISPRdisco,” the program identified more than 1000 different unexplored CRISPR systems, which the researchers whittled down to 50 candidates for Gersbach’s laboratory to generate and further engineer.

    Those CRISPR systems were then tested in human cells for their abilities as gene repressors and activators as well as genetic and epigenetic editors. While four systems stood out for their individual successes, one was particularly noteworthy for its versatility across the board. Called SubCas9, the promising CRISPR component was found in Streptococcus uberis, bacteria commonly found in dairy cows that are also used in some human probiotic products.

    The researchers are excited about SubCas9 for several reasons. It’s smaller than the traditionally used Cas9 DNA molecular scalpel, meaning it could more easily be loaded into delivery systems that efficienctly transfer cargo to human tissues. It can also target different genetic sequences than its original counterpart. Whereas the most commonly used Cas9 works at genomic targets adjacent to the DNA sequence “GG,” the new system works at sites neighboring “AATA” or “AGTA” patterns.

    “GG is a fairly common DNA sequence, but if you really need to target a specific base pair of DNA and there isn’t a GG nearby, then you need an alternative option,” said Gabe Butterfield, the postdoctoral fellow in the Gersbach Lab who co-led the work with Duke biomedical engineering PhD student Dahlia Rohm. “This system can give researchers flexibility for using different Cas9s when they need to be really precise with their target site selection.”

    Last but certainly not least, S. uberis is not commonly found in people, in contrast to bacteria species from which the more common Cas9 proteins have been isolated. This means that most people’s immune systems would not recognize SubCas9 from a previous natural exposure if it were to be used in a therapeutic application.

    Moving forward, the researchers are working to see if SubCas9 evades preexisting immunity as they expect and are testing incorporating it into a number of cell and gene therapies. They may also dip back into the massive bacterial metagenomic databases that are now available to find more CRISPR systems to investigate.

    “Besides potential for therapeutic applications, we also appreciate that bacteria that have adapted to diverse habitats harbor effectors better suited for various kinds of hosts, with much potential for discovery of systems more suited for plants, livestock and environmental applications,” said Barrangou.

    “We’ve been collaborating for several years now, and it’s been a really fruitful meshing of Duke’s biomedical abilities and NC State’s agricultural and microbiological expertise,” Gersbach said. “All of our work over the past decade has been made possible by investments from the National Institutes of Health to expand science’s genomic and epigenomic editing toolbox.”

    This work was supported by National Institutes of Health (U01AI146356, UM1HG012053, R01MH125236, RM1HG011123), the National Science Foundation (EFMA-1830957), DARPA (HR0011-19-2-0008), the Paul G. Allen Frontiers Group and the Open Philanthropy Project.

  • Bacterial ‘jumping genes’ can target and control chromosome ends

    Bacterial ‘jumping genes’ can target and control chromosome ends

    Transposons, or “jumping genes” — DNA segments that can move from one part of the genome to another — are key to bacterial evolution and the development of antibiotic resistance.

    Cornell University researchers have discovered a new mechanism these genes use to survive and propagate in bacteria with linear DNA, with applications in biotechnology and drug development.

    Researchers show that transposons can target and insert themselves at the ends of linear chromosomes, called telomeres, within their bacterial host. In Streptomyces — historically one of the most significant bacteria for antibiotic development — they found that transposons controlled the telomeres in nearly a third of the chromosomes.

    “This is a big part of their biology,” said senior author Joseph Peters, professor of microbiology. “Bacteria are like these little tinkerers. They’re always collecting these mobile DNA pieces, and they’re making new functions all the time — everything in antibiotic resistance is really about mobile genetic elements and almost always transposons that can move between bacteria.”

    With some technologies not available even five years ago, the researchers identified several families of transposons in cyanobacteria and Streptomyces that, using different mechanisms, can find and insert themselves at the telomere, with benefits for the transposon and their bacterial host. For one, inserting at the end of the chromosome helps the transposon avoid genes for the cell’s core functioning, which reside in the middle of the chromosomes; transposons that can target the ends are less likely to disrupt an essential function or cause cell death.

    “If you could target the end, you’re less likely to knock out something that the host wants, and then these ends, by various systems, are transferred between cells,” Peters said. “For any element to survive — a transposon, bacteria — they really need to be able to do those two things: They need to not cause too much damage, and they need a way to move to new hosts. By inserting into the telomeres, they’re able to do both.”

    Transposons have been found clustered at the chromosome ends in eukaryotic cells, but this is the first time it’s been documented in bacteria with linear chromosomes, and the researchers found that bacterial transposons (versus eukaryotes) use unique mechanisms to control the telomeres.

    Transposons are usually flanked by protein-binding sequences that indicate where to excise the DNA element and move it to a new location. In Streptomyces, researchers observed that the transposons at the telomeres were one-sided, with a traditional transposon sequence on one end with the other end being the telomere. This functionally allows the transposon to be the telomere, making it essential to the cell generally.

    “What it lets them do is become essential to the host, because they now control the telomere, and if the element got deleted along with this system, the host would die,” Peters said.

    The researchers found one subfamily of telomere-targeting transposons that coopted a CRISPR system — normally used by bacteria to defend against viruses — to target and insert itself into the chromosome ends. This process is further confirmation of previous research in Peters’ lab that found transposons using CRISPR systems to move around genomes, opening the potential for a new gene-editing tool that would allow for larger sections of DNA to be inserted than the now widely used CRISPR-Cas9.

    “The transposons keep grabbing these systems and coopting them in different ways,” Peters said. “In this paper, we explained a new one of these elements using a CRISPR-Cas system to target the telomeres.”

    The insights — especially into Streptomyces, which is difficult to manipulate in the lab and accounts for the discovery of many of our antibiotics — could prove useful for drug development, as transposons drive bacterial evolution and may direct researchers to new antibiotics and other useful products encoded on these transposons.

    “Most of life on the planet is microbes, and specifically bacteria,” Peters said. “We want to understand how these living organisms function, but then we want to see how we can use these systems for the betterment of humankind.”

    Co-authors include postdoctoral researchers Shan-Chi (Popo) Hsieh and Michael T. Petassi; doctoral student Richard Schargel; and partners at the University of Geneva, Orsolya Barabas and Máté Fülöp.

    The research was supported with funding from the National Institutes of Health and the European Research Council.