Nature Biotechnology, Published online: 08 April 2026; doi:10.1038/s41587-026-03087-3
Sequence Display maps protein variant activities to a sequencing-based readout.
New Single‑Cell Platform Tracks RNA and Protein in Immune Signaling
A new single‑cell sequencing method is giving researchers a clearer view of how immune cells actually behave—capturing not just what they plan to do, but what they are doing in real time. The platform, called CIPHER‑seq, measures RNA and proteins simultaneously inside the same immune cell, exposing gaps between genetic intent and functional output that have long complicated studies of cytokine signaling. The work, titled “CIPHER-seq enables intracellular multimodal profiling of cytokine responses in single immune cells,” appears in Scientific Reports.
Single‑cell RNA sequencing has reshaped immunology by revealing which genes are switched on across thousands of cells at once. But RNA alone can be misleading, especially for cytokines. However, RNA is only a set of instructions; proteins carry out the action. And for cytokines, RNA levels often fail to predict how much protein a cell actually produces. “In immune cells, RNA and protein don’t always rise and fall together,” said co‑senior author Emiliano Cocco, PhD, an assistant professor of biochemistry and molecular biology at the Miller School.
CIPHER‑seq (Cytokine Intracellular Protein High-throughput Expression with RNA-sequencing) was designed to close that gap. Developed by researchers at the Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine, together with collaborators at UCSF and the Helen Diller Family Comprehensive Cancer Center, the method gently preserves cells and captures multiple molecular layers at once. From a single immune cell, CIPHER‑seq can quantify genome‑wide RNA, surface proteins, intracellular proteins, and cytokines that have not yet been released—creating a more complete snapshot of immune activity than RNA‑only approaches.
“RNA gives us clues about where a cell is headed,” said co‑senior author Justin Taylor, MD, a Sylvester physician-scientist. “Proteins show us where it actually arrives, and this clearer picture could help scientists design better immunotherapies and help clinicians predict which patients are most likely to benefit from them.”
The team validated the platform by stimulating peripheral blood mononuclear cells (PMBCs) and tracking their responses. According to the study, CIPHER‑seq captured robust induction of key cytokines—including interferon‑gamma and tumor necrosis factor—while also resolving metabolic remodeling during activation. Importantly, the method revealed the timing of these events: RNA signals rose first, followed by delayed but consistent protein accumulation. First author Avni Bhalgat, PhD, described it as “seeing the plan before the action. Cytokines help determine whether immune cells attack cancer, ignore it, or even help tumors grow.”
The researchers also compared CIPHER‑seq with standard single‑cell workflows and found a notable difference: cells processed with CIPHER‑seq showed far fewer mitochondrial stress signatures. Some existing protocols inadvertently damage cells during preparation, triggering artificial stress responses. By reducing these artifacts, CIPHER‑seq provides a cleaner readout of immune behavior.
The authors emphasize that this multimodal view is especially valuable for studying cancer, inflammation, and treatment resistance—contexts where cytokine timing and protein abundance can shape therapeutic outcomes. “The platform helps us move beyond inference and toward understanding how immune responses truly unfold—one cell at a time,” Taylor added. By tracking RNA and protein together, CIPHER‑seq moves researchers beyond inference and toward a step‑by‑step understanding of how immune responses unfold.
The post New Single‑Cell Platform Tracks RNA and Protein in Immune Signaling appeared first on GEN – Genetic Engineering and Biotechnology News.
U.S. Tumor Testing Suffers Under Access Disparities
Genomic testing for advanced cancer in the U.S. is hampered by unequal access across demographic groups and needs targeted policy solutions, researchers report.
Next-generation sequencing (NGS) was not carried out for the five most prevalent types of solid tumor among thousands of patients studied, the team revealed in JAMA Network Open.
And among those who did undergo this testing, wait time significantly varied according to race, insurance status, and practice setting.
“Our findings highlight the underrepresentation of certain patient demographics in tumor genomic profiling, revealing disparities in access to standard-of-care diagnostic modalities,” reported researcher Chadi Hage Chehade, MD, from the University of Utah, and coworkers.
“These results emphasize the need for healthcare policies to mitigate these gaps.”
Precision oncology has defined a new era in cancer treatment, enabling clinicians to tailor care based on the specific clinicogenomic features of a patient’s tumor, enabling more effective and less toxic treatment strategies.
NGS has emerged as a transformative technology, enabling comprehensive genomic profiling and uncovering alterations for targeted therapies.
To examine equity of care in the field, the researchers studied electronic health record data for patients with common advanced or metastatic cancers that spanned over 800 U.S. community and academic sites across the U.S. between 2018 and 2022.
The team examined time to first NGS testing and frequency of testing for 63,294 patients, including those with metastatic breast (19.1%), prostate (6.9%), pancreatic (9.7%), colorectal (21.6%), and non–small cell lung cancer (42.7%, NSCLC).
The median age in the group was 68 years and 53.7% was female. In terms of ethnicity, 2.7% were Asian, 10.0% were Black, 6.0% were Hispanic, 61.0% were White, and 20.3% were other races and ethnicities.
The frequency of testing increased over the four-year span across all cancer types, but by the final year of study up to 40% to 50% of patients were still not receiving NGS testing.
Results showed there were differential rates of testing and longer waiting times to NGS testing in some groups.
Patients with lower socioeconomic status (SES), non-Hispanic Black or Hispanic patients, those covered by Medicare, Medicaid, or other government programs, and those treated at an academic practice setting were significantly less likely to be tested in some of cancers than patients with high SES, who were non-Hispanic White, those covered by a commercial health plan, or those treated in community practice, respectively.
Among specific cancers, Hispanic patients were significantly less likely to be tested in metastatic breast or prostate cancer, and non-Hispanic Black patients were less likely to receive NGS in advanced NSCLC, metastatic colorectal or metastatic pancreatic cancer.
The findings highlight the need to improve access to standard-of-care diagnostic modalities and serve as a call to improve NGS testing rates nationwide, said Igor Makhlin, MD, in an accompanying Commentary article.
“While the accelerating pace of research and AI-driven technology is poised to herald the next generation of discoveries that translate into greater survival for patients with cancer, we cannot ignore the increased burden to stay up to date, largely born by community oncologists who manage a wide gamut of solid and liquid cancers,” he maintained.
“Creation and adoption of innovative strategies to support clinicians in implementing breakthrough advances into their practice regardless of zip code, practice site, or other factors will require a concerted effort by all relevant stakeholders, but closing this gap in GCC is absolutely necessary. Our patients are depending on us.”
The post U.S. Tumor Testing Suffers Under Access Disparities appeared first on Inside Precision Medicine.
STAT+: Many cancer patients don’t get genomic tests to guide treatment, study finds
For some advanced cancers, sequencing the tumor genome should be one of the first steps patients and physicians take. But a new study finds that many patients never receive genomic testing and so never get the chance to know if they might have benefitted from newer, more targeted therapies.
The study, published on Tuesday in JAMA Network Open, examined how many patients diagnosed with one of five different metastatic cancers received genetic sequencing for the cancers. For most cancers in the study, roughly half of patients in the cohort received genetic sequencing. Patients with low income, Medicare or Medicaid coverage, and Black or Hispanic race or ethnicity were also less likely to receive sequencing.
Cancer medicine and research have made enormous progress over the last few decades. The overall five-year survival rate has pushed up to 70% as of 2026, and the five-year survival rate for metastatic cancer has doubled since the 1960s. That’s in large part thanks to advances in medicines and technologies that can help treat cancer, like targeted therapies that work by exploiting key cancer mutations.
CRISPR at 25: The Past, Present, and Future of Genome Editing
Panelists:
Rodolphe Barrangou, PhD
North Carolina State University
Panelist
Rodolphe Barrangou, PhD
Rodolphe Barrangou, PhD, is the T. R. Klaenhammer Distinguished Professor at North Carolina State University, where he leads the CRISPR Lab. Rodolphe spent nine years at Danisco and DuPont, where he made seminal contributions in the functional characterization of CRISPR as a microbial immune system. He has been at NC State since 2013.
For his CRISPR work, Rodolphe has received several international awards, notably the Canada Gairdner International Award, and has been elected to the National Academy of Sciences, the National Academy of Engineering, and the National Academy of Inventors. Rodolphe is a scientific co-founder of Intellia Therapeutics, Locus Biosciences, TreeCo, Ancilia Biosciences, and CRISPR Biotechnologies, and an advisor to Inari and the IGI. He is also the founding Editor in Chief of The CRISPR Journal (published by Mary Ann Liebert, Inc., a Sage partner), which launched in 2018.
Rodolphe holds a degree from Paris Descartes University and a PhD in functional genomics from NC State.
Broadcast Date:
- Time:
It has been almost 25 years since the acronym “CRISPR” was first coined. Since then, CRISPR has become a household word, a star of books and films, and a Nobel Prize–winning discovery. This powerful and disruptive genome editing technology has transformed countless fields, including gene therapy, xenotransplantation, de-extinction and agbiotech. Researchers continue to build on the CRISPR chassis, devising new platforms for bespoke genome editing. But major questions remain around clinical safety, commercial development, ethical deployment, and regulatory oversight.
In the first of a new series of GEN Keynote Webinars, Professor Rodolphe Barrangou, PhD (North Carolina State; EIC, The CRISPR Journal) offers a front-row perspective of the CRISPR revolution, the seminal advances, clinical highlights, and rising applications. Almost two decades ago, Barrangou provided the first experimental demonstration of the functional role of CRISPR. With numerous advisory and entrepreneurial activities in the gene editing space, Barrangou is the ideal guide to discuss CRISPR’s progress in the clinic; the state of the CRISPR toolbox; and the regulatory roadblocks and ethical challenges that will shape the application of CRISPR in agbiotech, germline editing, and other arenas.
Registration for this GEN Keynote Webinar is free. Following this live presentation, Dr. Barrangou will answer audience questions.
Produced with support from:
The post CRISPR at 25: The Past, Present, and Future of Genome Editing appeared first on GEN – Genetic Engineering and Biotechnology News.
Low-Cost, Single Sample Blood Test Detects Different Cancers, Liver Disorders, and Other Diseases
UCLA scientists have developed a simple and cost-effective blood test that, in early studies in more than 1,000 people, showed promise in detecting multiple cancers, various liver conditions, and organ abnormalities simultaneously.
The new method, called MethylScan, works by analyzing cell-free DNA (cfDNA), tiny fragments of genetic material released into the blood when cells die. Because cells from every organ shed DNA into the bloodstream, cfDNA carries molecular signals that reflect what is happening throughout the body.
The researchers say MethylScan could represent a powerful and more affordable approach to early disease detection and comprehensive health monitoring. “Early detection is crucial,” said research lead Jasmine Zhou, PhD, a professor of pathology and laboratory medicine and investigator at the UCLA Health Jonsson Comprehensive Cancer Center. “Survival rates are far higher when cancers are caught before they spread. If you detect cancer at stage one, outcomes are dramatically better than at stage four.” Zhou is senior author of the team’s published paper in PNAS, titled “Toward the simultaneous detection of multiple diseases with a highly cost-effective cell-free DNA methylome test.”
“When cells die, they do not simply vanish; they leave behind molecular traces, including cell-free DNA (cfDNA) in the blood stream,” the authors wrote. “cfDNA is a mixture of DNA fragments released from various organs, offering valuable insights into the health of these organs.” Zhou added, “Every day, 50 to 70 billion cells in our body die. They don’t just disappear, their DNA goes into the bloodstream. That means we already have information from all our organs circulating in the blood.”
The idea of using blood to detect cancer, sometimes called a liquid biopsy, isn’t new. Some tests already look for mutations in tumor DNA to screen for certain cancers. But those tests often focus on a limited number of genetic changes and can be expensive, in part because they require deep sequencing to detect faint tumor signals.
Instead of searching for mutations, the UCLA team examined DNA methylation, chemical tags attached to DNA that help regulate gene activity. Methylation patterns differ by tissue type and can change when cells become cancerous or diseased. “Unlike an individual’s genome, which remains largely stable across tissues and over time (except for rare somatic mutations), the DNA methylome is tissue-specific and dynamically changes with the tissue’s disease status,” the team continued.
“DNA methylation reflects the health status of a tissue,” said co-corresponding author Wenyuan Li, PhD, a professor of pathology and laboratory medicine at UCLA and co-corresponding author of the study. “It’s a very informative signal.”
The challenge is that most cell-free DNA in the bloodstream doesn’t come from tumors or injured organs. About 80% to 90% originates from normal blood cells. That creates background noise, making it difficult and costly to detect the relatively rare fragments that might signal early cancer. “A major challenge in using the cfDNA methylome for disease detection is the high cost of sequencing,” the team stated. “In healthy individuals, about 85% of cfDNA originates from blood cells, creating substantial background noise that can obscure cfDNA from tumors or diseased organs.” And as the authors further pointed out, “Current cfDNA methylation assays primarily focus on single clinical indications by targeting specific genomic loci.”
For their newly reported research the team built on past work to develop a technique to remove much of the background DNA before sequencing. Using specialized enzymes, they selectively cut away unmethylated DNA fragments that largely come from blood cells. By designing a genome-wide hybridization panel, the captured DNA fragments are enriched for methylated DNA from solid organs, including those that are potentially diseased. “The MethylScan method is a targeted methylation assay that combines Methylation-Sensitive Restriction Enzymes (MSRE) digestion with a custom panel to enrich hypermethylated cfDNA from tissues beyond blood, enabling cost-effective detection of multiple diseases,” they wrote.
By removing the noise, the researchers say they can dramatically reduce the amount of sequencing needed, lowering costs while maintaining sensitivity. Achieving an effective sequencing depth of 300× per sample requires only 5 Gb of data, which would cost less than $20 if the price per gigabase is under $4.
To test the accuracy of MethylScan, the researchers analyzed blood samples from 1,061 people, including patients with liver, lung, ovarian and stomach cancers; individuals with liver diseases such as hepatitis B, hepatitis C, alcohol-related liver disease, and metabolic-associated liver disease; people with benign lung nodules; and healthy participants. Machine learning algorithms were then applied to analyze the complex methylation data.
For multi-cancer detection, the test achieved a high level of overall accuracy. At a specificity of 98%, meaning few false positives, it detected about 63% of cancers across all stages and roughly 55% of early-stage cancers. The test also performed well in liver cancer surveillance among high-risk individuals, including those with liver cirrhosis or HBV, detecting nearly 80% of cases at a specificity of just over 90%, meaning a less than 10% false positive rate.
Beyond simply detecting cancer, the methylation patterns helped identify where in the body a signal was coming from, known as the tissue of origin. “Being able to trace signals back to their source is important because a positive blood test needs to be followed by imaging or other diagnostic procedures directed at the right organ,” said Li.
MethylScan can work like a health radar for the body. By reading DNA signals in the blood, it can tell when specific organs, such as the liver or lungs, are under stress or damaged, even without knowing the disease in advance. The researchers also showed that the blood test could distinguish between different types of liver disease, including viral hepatitis and metabolic-associated liver disease. It correctly classified about 85% of patients, suggesting blood-based DNA testing could reduce the need for invasive liver biopsies.
Although larger prospective trials will be needed to confirm its performance in real-world screening, Zhou said the work represents an important step toward a single, affordable blood assay that can detect a broad spectrum of diseases earlier and more comprehensively than current methods allow.
“Because cfDNA in blood originates from multiple organs, and MethylScan captures a broad spectrum of robust hypermethylation markers, this assay has the potential to detect a variety of diseases, provided that appropriate training cohorts are available,” the investigators stated. “This versatile approach enables affordable, wide-ranging cfDNA tests that can identify various health conditions simultaneously, with the potential to transform early disease detection and health monitoring across diverse clinical settings.
Zhou added, “This study demonstrates that blood-based methylation profiling can deliver clinically meaningful information across multiple diseases. It’s an exciting advancement that brings us closer to realizing the dream of a single assay for universal disease detection.”
The post Low-Cost, Single Sample Blood Test Detects Different Cancers, Liver Disorders, and Other Diseases appeared first on GEN – Genetic Engineering and Biotechnology News.
Top 5 Firms Engineering Healthcare in the CNS Space
Central nervous system (CNS) treatments are having a major comeback. These five precision medicine players plan to ride the resurgence.
After a decade of stagnation, the CNS space is seeing a revival in sales and R&D spending as the market was last year projected to surpass $80 billion for the first time since 2013 and hit around $127 billion.
Recent landmark approvals have brought attention back to the CNS, including the U.S. Food and Drug Administration (FDA)’s greenlight of Eisai/Biogen’s lecanemab (Leqembi) for the treatment of Alzheimer’s disease in 2023, and the FDA approval of Bristol-Myers Squibb’s schizophrenia treatment xanomeline/trospium chloride (Cobenfy) in 2024.
At the same time, Johnson & Johnson’s depression treatment, esketamine (Spravato), is on its way to blockbuster status, showcasing the growth potential of the CNS market.
These successes accompany an emerging shift in psychiatry clinical trials from subjective rating scales to more objective endpoints, including digital and physiological measures, with the potential to better tailor treatments to a patient’s biological makeup.
Startups and scaleups are attracting increasing investor attention for their potential to change the way we treat CNS conditions. Check out our list of the most exciting companies that have netted the biggest investor dollars.
1. Aerska
Founded: 2025 | Headquarters: Dublin, Ireland
Aerska’s name is derived from an Irish proverb stating that people survive in each other’s shelter, emphasising the strength of its team.
This team includes co-founder Jack O’Meara, previously co-founder of the liver-focused RNA interference (RNAi) biotech Ochre Bio, who is driven by the experience of loved ones suffering from Alzheimer’s disease.
Aerska is developing RNAi therapies for neurodegenerative conditions, including Parkinson’s and Alzheimer’s disease.
While there are already FDA-approved RNAi therapies on the market, such as Alnylam’s patisiran (Onpattro), these are typically focused on liver and cardiometabolic conditions rather than the CNS.
Aerska’s technology consists of antibody “brain shuttles” that bind to proteins on the blood-brain barrier (BBB). They then carry a payload RNA into the brain.
The payload, which is designed based on data-driven patient stratification and disease biomarkers, then silences specific genes driving the disease.
Aerska has already raised $60 million since its launch, including a $21 million seed round in October 2025 and a $39 million Series A round in February 2026, co-led by EQT Life Sciences and age1.
The company, which has research operations in the U.K., is using the latest funding to drive its pipeline programs toward clinical testing.
2. Beacon Biosignals
Founded: 2019 | Headquarters: Boston, Massachusetts, U.S.
Beacon Biosignals was co-founded by a team including its CEO—MIT neuroscientist Jacob Donoghue, MD, PhD—and its CTO, the machine learning researcher Jarrett Revels.
Boasting more than 100 employees, the company’s goal is to provide objective biomarkers in drug development that neurology and psychiatry have traditionally lacked compared with other areas of precision medicine.
Its FDA-cleared Waveband device measures the brain’s activity, known as electroencephalography (EEG), while patients sleep at home. The EEG data is then stored, quality-controlled, and fed into AI models that can guide the design of clinical trials.
For example, Beacon’s EEG data can identify patients with Alzheimer’s disease who have worse outcomes and might need a more targeted treatment or a different clinical trial than other patients.
Beacon raised $27 million in a Series A round in 2021 and an oversubscribed Series B round worth $86 million in November 2025.
The B round, which included investors such as Innoviva, Google Ventures, and Nexus NeuroTech, will help the startup to accelerate the discovery of neurobiomarkers and broaden clinical adoption of the technology.
Beacon acquired the French sleep monitoring company Dreem in 2023 to access its monitoring data and headband technology. Beacon then acquired the Ohio-based CleveMed in April 2025 to harness technology measuring breathing, oxygen, and other signals.
3. Brainomix
Founded: 2010 | Headquarters: Oxford, U.K.
Brainomix was founded by a team including CEO Michalis Papadakis, PhD, who was scientific director of the preclinical stroke lab at the University of Oxford.
Brainomix is dedicated to speeding up patient care in cases of stroke, where speedy treatment is key.
Brainomix’s flagship product, Brainomix 360 Stroke, is designed to harness AI to interpret brain scans and detect blood clots in patients with stroke, speeding up clinical decision-making.
The product involves a group of tools that automatically analyze images, including results from computed tomography (CT), CT angiography, magnetic resonance imaging (MRI), and CT perfusion.
Brainomix’s technology doubled the rate of thrombectomy treatment in patients with stroke and reduced hospital triage and transfer delays, according to a 2025 study.
The University of Oxford spinout is at a commercial stage, with operations in more than 20 countries, and is expanding into the U.S.
Brainomix raised a $21.2 million Series B round in 2021 and extended its Series C round from $6.5 million in March 2025 to $25.4 million in February 2026, with leading investors including Parkwalk Advisors and Hostplus. The proceeds will fuel the company’s expansion into the U.S. market.
Brainomix has also partnered with heavyweights, including Nvidia, Boehringer Ingelheim, Medtronic, and GE Healthcare.
Brainomix also has a product dedicated to disease monitoring in pulmonary fibrosis.
4. Circular Genomics
Founded: 2021 | Headquarters: San Diego, California, U.S.
Circular Genomics was spun out of the University of New Mexico, with its founders including CSO Nikolaos Mellios, PhD, and Alexander Hafez, PhD.
The company later moved its headquarters from Albuquerque to San Diego in March 2025 to access scientific and operational know-how from Eli Lilly at Lilly Gateway Labs.
Circular Genomics aims to equip medical professionals with a blood test to detect CNS conditions early, in addition to stratifying and guiding the treatment of patients.
Its technology involves using a polymerase chain reaction (PCR) test of a patient’s blood sample to screen for specific circular RNA molecules produced in the brain that can cross into the blood and be measured as a biomarker of disease in the CNS.
Commercially launched in 2024, Circular Genomics’ MindLight SSRI Antidepressant Response Test predicts whether a patient will benefit from common antidepressants called SSRIs with around 77% accuracy. This is designed to predict a patient’s most suitable antidepressants without needing months of trial-and-error approaches.
The company is applying its technology in Alzheimer’s disease, where the approvals of disease-modifying therapies such as Leqembi have led to demand for tests that can detect the disease at earlier stages than traditional tests.
Circular Genomics raised $15 million in a Mountain Group Partners-led Series A round in December 2025 to finance the development of its technology and expansion of its technology in Alzheimer’s disease.
The company also has its sights on other CNS conditions, including multiple sclerosis and Parkinson’s disease.
5. Omniscient Neurotechnology
Founded: 2019 | Headquarters: Sydney, Australia
Omniscient (o8t)’s founders include CMO Michael Sughrue, MD, a neurosurgeon aiming to improve anatomy maps for other surgeons, and machine learning expert Stephane Doyen, PhD.
o8t’s FDA-approved product Quicktome involves using a patient’s MRI brain scans and AI models to map out a patient’s brain circuitry. These maps, accessible from an electronic tablet, can guide surgery to minimize the risk of brain damage compared to using a generalized anatomical diagram.
Quicktome is already in use at major hospitals around the world, including major centers in the U.S. Its partners include U.S. surgical support firm META Dynamic and the U.S. medical device innovation center, The Jacobs Institute.
o8t has raised more than $60 million, and bagged $14 million (AUD 20 million) in January 2026 as part of a Series D round targeted to reach $25 million (AUD 36 million). The round was led by Australia’s National Reconstruction Fund (NRFC) and OIF Ventures, with the aim of keeping the company based in Australia.
The funding is earmarked to fuel the development and commercialization of Quicktome, and grow o8t’s Australian workforce by more than 40. The company also has operations in Atlanta, Georgia, U.S.
o8t also plans to expand the technology into high-growth markets, including brain computer interface targeting, stroke and traumatic brain injury.
Jonathan Smith, PhD, is a freelance science journalist based in the U.K. and Spain. He previously worked in Berlin as a reporter and news editor at Labiotech, a website covering the biotech industry. Prior to this, he completed a PhD in behavioral neurobiology at the University of Leicester and freelanced for the U.K. organizations Research Media and Society of Experimental Biology. He has also written for medwireNews, Biopharma Reporter, and Outsourcing Pharma.
The post Top 5 Firms Engineering Healthcare in the CNS Space appeared first on Inside Precision Medicine.
Author Correction: Real-world clinical utility of tumor whole-genome sequencing in solid cancers
Nature Medicine, Published online: 02 April 2026; doi:10.1038/s41591-026-04379-6
Author Correction: Real-world clinical utility of tumor whole-genome sequencing in solid cancers
Bacteria Defense Insights Could Revolutionize Genetic Editing
A computer program that can predict which genes help bacteria to defend themselves against viruses could lead to the next generation of precision genetic engineering tools.
The artificial intelligence model recognizes genetic sequences involved in defenses that act against bacteriophages—viral invaders that infect bacteria.
These anti-viral immune systems have already been repurposed into powerful gene-editing technology, such as CRISPR-Cas, that enable DNA sequences to be precisely cut, modified, or deleted within an organism.
The DefensePredictor tool, outlined in Science, is available as an open-source tool to enable the discovery of more prokaryotic immune systems.
“Identifying new antiphage defense systems may yield the next generation of precision molecular tools while also shedding important light on the ongoing arms race between bacteria and phages,” said MIT-based molecular biologist Michael Laub, PhD, and co-workers.
Intense selective pressure to evade or survive infection has driven the evolution of numerous antiphage defense mechanisms, including restriction enzymes and the CRISPR-Cas systems.
While antiphage immunity genes often cluster into “defense islands” in prokaryotic genomes, this does not always occur and many systems are dispersed or carried on mobile elements such as plasmids, prophages, and transposons.
In an attempt to create a model to identify antiphage proteins, Laub and team first looked at around 17,000 genomes of prokaryotic organisms.
They labelled homologs of known defense and nondefense genes and built representations of the proteins coded by these genes as well as their four nearest neighbors on the genome.
DefensePredictor was trained through this to distinguish whether a gene was involved in defense systems.
After performing well in silico, it was tested on 69 diverse Escherichia coli genes and identified 624 different proteins that it confidently predicted were involved in defense, including 154 that shared no detectable homology to known defense proteins.
Nearly half of the defense proteins identified were not encoded in plasmids, prophages, or defense islands, showing that the model was able to identify systems in a wide range of genomic contexts.
Of 94 predicted genes tested in the lab, 42 provided protection against at least one of 24 phages tested, giving a validation rate of around 45%.
Fifteen protein domains across these 42 systems had not previously been validated as defensive, suggesting new immune systems remain undiscovered.
Expanding the predictive capacity of DefensePredictor beyond E. coli to 1000 diverse prokaryotic genomes revealed more than 5000 predicted defense proteins that were not clear homologs of those already known.
Another Science research article in the same issue of the journal also showed how AI could uncover unexplored diversity in bacterial immunity.
Ernest Mordret, PhD, from the Pasteur Institute, and co-workers demonstrated how deep-learning frameworks could lead to the large-scale discovery of antiphage and a vast atlas of bacterial antiviral immunity.
The team developed three complementary deep-learning models to predict antiphage proteins by leveraging genomic context (ALBERTDF), amino acid sequence (ESMDF), or both (GeneCLRDF).
Twelve newly predicted antiphage systems were then experimentally validated in Escherichia coli and Streptomyces albus.
When applied to more than 30,000 bacterial genomes, the models predict 2.39 million antiphage proteins, 85% of which had no previously known link to immunity, corresponding to approximately at least 23,000 predicted antiphage operon families.
All predictions have been made freely available through an interactive antiphage atlas.
“We developed deep learning models to predict antiphage systems,” the authors summarized.
“These methods extract cues about the “defensiveness” of a protein from two seemingly orthogonal sources: its genomic context across thousands of genomes, and its own amino acid sequence.
“By combining these complementary signals, we move from a fragmented, incomplete view of bacterial immunity toward a more resolved and quantitative understanding of its repertoire.”
The post Bacteria Defense Insights Could Revolutionize Genetic Editing appeared first on Inside Precision Medicine.
AGBT 2026 Recap: NGS Big Bets and Spatial’s Rising Momentum
For most people, February in Florida means school breaks, water parks, and trips to Disney. But for the genomics community, that combination means the season of big announcements as the Advances in Genome Biology and Technology (AGBT) meeting—which has earned a reputation for breaking field-shaping news—takes place. The meeting is packed not only with technology announcements in the sponsor suites, but also with scientific talks in the sessions to showcase how the technology is being used to address new biological questions.
The first piece of news was the weather. Not in Orlando, although it was uncharacteristically cold. But for the nail-biting attendees traveling from the Northeast who were watching the prediction of inches of snow grow with each forecast. Those of us who bumped up our flights to beat the storm and make some of the last flights out of the Northeast were lucky enough to arrive in time for the opening session.
The meeting was, as usual, a constant stream of announcements and advancements. Some grumbled that the meeting was slower than usual, but that seems unfounded to this attendee. One NGS company launched two new instruments just two years after its first instrument; the very first 3D spatial instrument was launched commercially, while other spatial news showed a maturing of the field. Many attendees could not stop talking about some of the research presented—including the “bat talk” given by Emma Teeling, PhD, from the University College Dublin, during the opening session (which was officially named, “Bats: new models of extended healthspan and disease resistance”).
Also included in the opening session was an award presentation to Eric Green, MD, PhD, the former National Human Genome Research Institute (NHGRI) director. Green was sorely missed at last year’s 25th anniversary meeting because he (and many others working at the NHGRI and NIH) were not in attendance due to travel restrictions on government staff. A short time later, Green’s 15-year stint at the NHGRI was terminated, becoming the first of multiple NIH directors to be ousted by the Trump administration. But now, Green has completed his transition from government work to a new role as Illumina’s CMO. However, this transition also means that Green can no longer serve as a program committee co-chair for the AGBT General Meeting, which he has done for over 25 years, making the Distinguished Service Award a fitting end to his tenure. The meeting, Green noted on LinkedIn, has “always been about charting a course for the future.”
Sequencing steals the spotlight
Next Generation Sequencing (NGS) dominated the buzz at the meeting this year. Ultima Genomics made news just before the meeting started, revealing two new instruments: the UG 200 single-wafer and the UG 200 Ultra dual-wafer. Both boxes are less expensive and have higher throughput than the original iteration—the UG 100. Gilad Almogy, PhD, Ultima’s CEO, told GEN that the UG 200 series is more mature because it has been developed through “a ton of learning of how [the UG 100] performed in the field.” But it was a quick learning cycle, as the UG 100 was only launched in 2024.
In contrast, AGBT attendees had to wait patiently for the most anticipated news from the meeting: Roche filling in missing details about its sequencing by expansion (SBX) nanopore instrument, the Axelios. The company’s lunch talk was scheduled on the last day of the meeting, with many people rolling their luggage around in preparation to depart. Roche had already announced the cost of the Axelios instrument at $750,000, but the pricing for the consumables and the launch date remained unknown. Roche did offer some information, announcing a whole genome price of $150 (in duplex mode), a simplex price of $0.06 per million reads, and availability sometime this summer. A few new, notable discussion points were raised, including the length and complexity of the duplex sequencing library prep.
Illumina, the gold sponsor of the meeting, doubled down on its multiomics theme by presenting a complete multiomics workflow with a focus on the company’s longer read TruPath product. The kit, which uses an on-flow cell library preparation to obtain long-read insights, was first presented in 2024 as constellation mapped read technology. In short, the DNA spans multiple wells that are spaced under 100 nm apart. The DNA is fixed and then undergoes clustering and sequencing, connected throughout the DNA molecule. TruPath enables haplotyping, structural variant detection, and short tandem repeat analysis, which Steve Barnard, PhD, CTO of Illumina, said is “creating a new category of sequencing and giving the insights we need to diagnose patients.” The company emphasized TruPath’s ease of use during their talk by including a photo of Green using it at the bench and noting that it is so easy, even an executive can use it.
Element Biosciences did not wait until AGBT to reveal its latest innovation. The company announced its new high-throughput benchtop sequencing system, VITARI, in a webinar the week before the conference. The company noted that the instrument will begin shipping in the second half of 2026 and had a roadmap that included future multiomic capabilities.
In other industry news, Complete Genomics confirmed that it had entered into a definitive agreement to be acquired by Swiss Rockets AG, a Switzerland-based life sciences group. This move splits the company from Chinese ownership by MGI/BGI, and it will become a subsidiary of the Swiss life sciences group. Rade Drmanac, PhD, Complete’s co-founder and CSO, told GEN that this news allows the company to continue its focus on NGS instrumentation but also grow the focus into applications.
Expansion of spatial biology
Despite the wave of high-profile NGS announcements, spatial biology held its ground as a major focus at AGBT with updates reinforcing the technology’s maturity and expansion into new areas.
Vizgen announced updates on its MERSCOPE Ultra platform, including expanding its portfolio of predesigned panels and introducing a new customization capability. In addition, the company covered upcoming workflow innovations for upstream sample preparation and downstream bioinformatics. But perhaps the coolest update was the company’s work on organoids—a field where spatial analysis has proven challenging. Several characteristics of Vizgen’s platform are now enabling spatial analysis of organoids.
Singular Genomics’ new G4X Spatial Sequencer was on display in its suite, which the company launched the week before the meeting. At AGBT, the G4X platform was featured in a talk on SPOT-Met (Spatial Predictors of Tropism and Metastasis) by Jiwoon Park, PhD, from the lab of Christopher Mason, PhD, at Weill Cornell Medicine. SPOT-Met is a 1,000-tumor colorectal cancer program described as the largest colorectal cancer multimodal spatial initiative. “Population-scale spatial has arrived and is on center stage at AGBT 2026,” said Mason.
Attendees who visited the Stellaromics suite, which many did, judging from the activity, were encouraged to forget about 2D spatial and start thinking 3D. A Boston-based AGBT newcomer, Stellaromics is leading the 3D spatial wave with the launch of the first 3D commercially available spatial imager—the Pyxa. Although it may have been Stellaromics’ first time in Florida for the meeting, the company is led by genomics veteran Todd Dickinson, PhD (previously from Illumina, Bionano, Dovetail Genomics), who is no stranger to AGBT.
Last year, Bruker Spatial Biology established its place as a contender in the spatial world, just one year after the NanoString acquisition. This year, the company solidified its place as a leader by launching two products that it spoke about last year: CellScape (for spatial proteomics) and PaintScape (for visualization of the 3D genome). It also noted the mouse whole transcriptome for the CosMx Spatial Molecular Imager with 64 proteins. The technology was highlighted in multiple talks, including that of Miranda Orr, PhD, from Washington University, as she delved into the world of 3D reconstruction of neuropathology in the Alzheimer’s brain.
Although 10x Genomics is typically a top-tier sponsor at AGBT, the company was relatively quiet this year. However, it still managed to create buzz by delivering chocolate bars to each attendee’s hotel room stamped with a date: 4/18/26. Mid-April falls at the beginning of the American Association for Cancer Research (AACR) meeting, leaving something to look forward to.
Multiomics and more
BD Biosciences (now Waters) made a strong presence in the single-cell multiomics space with a focus on multimodal cellular profiling using its Rhapsody System. The new roadmap for this system piqued interest. In addition, the new hire of spatial veteran Luciano Martelotto, PhD, as director of global market development (single cell), working in the suite from morning until night, helped highlight the company’s place at the meeting.
Newcomer Syndex Bio introduced its mcPCR (methyl-copying PCR) platform, which enables copying of both DNA and methylation during amplification. Codetta Bio’s Concerto multiomic system, which detects DNA, RNA, and protein biomarkers in a single run, became commercially available (after being introduced at AACR last year), and the company spoke about the upcoming launch of new customizable panels for immunology and neuroscience.
In other innovative technologies, Gary Schroth, PhD, CSO of Cellanome, presented the company’s CellCage technology for the first time, which can study cells to understand their history and collect transcriptome data over time. Schroth showed a video of glial cells phagocytosing bacteria, and measuring the functional changes with the cells’ gene expression changes. Volta Labs announced the expansion of the capabilities of its Callisto platform, collaborations with Roche and Watchmaker
Genomics, and unveiled a growing pipeline of applications rolling out through 2026.
And that is the real takeaway from AGBT: innovation does not stop when people fly home. All of these announcements, made over four days, are significant advances and the excitement is palpable. But the truth is, innovators in the genomics field continue to push the boundaries all year long. For those of us who are passionate about genomics, we will look forward to seeing what they’ll unveil next year.
The post AGBT 2026 Recap: NGS Big Bets and Spatial’s Rising Momentum appeared first on GEN – Genetic Engineering and Biotechnology News.