Researchers have developed AI tool to rival the behemoth of Google’s Alphafold 3 in predicting 3D shapes in RNA using less data.
RNAbpFLow could shed new light on the RNA conformational dynamics that underpin diverse cellular processes.
It could also lead to novel RNA-based treatments, such as the messenger RNA vaccines used to prevent COVID-19.
The invention, by two computer scientists at Virginia Tech, generates all-atom RNA conformational ensembles for single-chain RNA monomers.
Unlike several existing deep-learning methods, it can do this without using evolutionary information or homologous structural templates.
The approach is outlined in Nature Methods and the researchers have made the training data, and code freely available.
Study first author Sumit Tarafder, a PhD student, flagged the importance of knowing the shape of an RNA so that it could be targeted.
“In the shape, there are pockets where a drug can attach,” he explained. “If you can’t predict the shape, your pockets are wrong—and the drug won’t work.”
Growing interest in RNA-based therapeutics has driven efforts to determine the 3D structures of RNA.
However, the intrinsic conformational flexibility of RNA presents major challenges when using methods such as X-ray crystallography, nuclear magnetic resonance spectroscopy, and cryo-electron microscopy.
Computer-based methods have emerged as an attractive alternative, but several of these approaches are constrained by the scarcity of RNA structural data in the Protein Data Bank.
While a growing number of methods based on Deep Learning have emerged, most are highly dependent on explicit evolutionary sequence information derived from multiple sequence alignments (MSA) or implicitly make use of homologous information learned by biological language models.
Tarafder and associate professor Debswapna Bhattacharya therefore developed RNAbpFlow, a sequence- and base pair-conditioned all-atom RNA 3D structure generation method based on SE(3)-equivariant flow matching model.
Doctoral student Sumi Tarafder (left) and Associate Professor Debswapna Bhattacharya explain a new AI method that rivals Google in decoding RNA, an approach that could help discover new treatments for disease [Tonia Moxley / Virginia Tech]
RNAbpFlow incorporates conditions on the nucleotide sequence and base-pairing information from three complementary base pair annotation methods to comprehensively capture canonical and noncanonical interactions.
By incorporating a nucleobase center representation that enables the optimization of angles of all rotatable bonds of nucleobases, it directly outputs all-atom RNA structures in an end-to-end fashion.
This bypasses the need for a post-hoc geometry optimization module, which is impractical in the context of large-scale sample generation.
Base pair-centric auxiliary-loss functions maximize the realization of canonical and noncanonical base-pairing interactions. This enables efficient generation of all-atom RNA conformational ensembles while explicitly modeling nucleobase orientation and flexibility.
Experimental results demonstrated that the introduction of base-pairing conditioning led to improved performance and accuracy connected to the quality of the base pairs.
In blind testing, RNAbpFlow produced a correct overall structure for 12 of 14 RNA targets, compared with eight out of 14 for AlphaFold 3, from Google DeepMind.
“We wanted to keep it simple and predict the structure from scratch, using just the sequence and the base pairs,” Tarafder said.
“The model starts from complete noise and, guided by those base pairs, folds into the right 3D shape.
“That’s the beauty of flow matching, and we can generate as many structures as you want, which lets us capture how the molecule actually moves.”
<![CDATA[Check out the pipeline updates from June!]]>
Posted on
<![CDATA[Why psychiatry’s growing “tribalism” harms care—and how clinicians, academia, and pharma can unite to tackle untreated mental illness.]]>
Posted on
<![CDATA[Track latest neuropsychiatry breakthroughs: phase 3 narcolepsy study launches, new depression and PTSD therapies advance, and FDA approvals reshape care.]]>
Researchers have identified a distinct set of anti-lipid antibodies that could improve the early diagnosis of Lyme disease while also helping distinguish patients who develop persistent symptoms after treatment from those who recover fully.
The study, published in Infection and Immunity, found that antibodies against specific phospholipids appear before conventional Lyme disease antibodies in some patients and persist in a subset of individuals with post-treatment Lyme disease (PTLD). The findings suggest these immune markers could eventually complement existing diagnostic tests and provide new clues about the biology underlying chronic symptoms.
Lyme disease, caused by the spirochete Borrelia burgdorferi, is the most common vector-borne infection in North America and Europe. Although most patients respond well to a two- to three-week course of antibiotics, an estimated 10% to 20% continue to experience fatigue, pain, cognitive impairment, and other nonspecific symptoms long after treatment. The biological basis for these persistent symptoms remains poorly understood, and clinicians currently lack objective biomarkers to identify affected patients.
Current diagnostic testing also has significant limitations. Standard two-tier serologic testing relies on antibodies directed against Borrelia proteins, but these antibodies often do not appear until several weeks after infection and can remain detectable long after the bacteria have been eliminated. As a result, existing tests have limited sensitivity during early infection and cannot reliably distinguish active infection from previous exposure or persistent post-treatment illness.
The new study builds on earlier work showing that B. burgdorferi scavenges lipids from its human host and incorporates them into its outer membrane. The researchers hypothesized that this unusual biology could trigger immune responses against host lipids that might serve as biomarkers of disease activity.
To test that idea, investigators analyzed serum samples from patients with acute Lyme disease and PTLD collected from two independent biobanks. Together, the cross-sectional and longitudinal cohorts allowed researchers to follow antibody responses from the day of diagnosis through one year after treatment.
Among numerous lipid targets examined, only three antiphospholipid antibodies were consistently elevated following infection. Two antibodies—anti-phosphatidic acid (αPA) and anti-phosphatidylserine (αPS)—were significantly elevated at diagnosis, including in patients with erythema migrans before they had seroconverted on conventional Lyme disease testing.
Longitudinal analyses showed that antibody levels followed distinct patterns over time. Both αPA and αPS peaked approximately three to six months after diagnosis before declining in most patients. However, αPS remained persistently elevated in a subset of individuals with PTLD, distinguishing them from healthy controls and from patients with autoimmune and chronic illnesses that often resemble post-treatment Lyme disease, including systemic lupus erythematosus, multiple sclerosis, fibromyalgia, long COVID, and chronic fatigue syndrome.
The researchers propose that persistent αPS elevations may reflect an abnormal immune response rather than ongoing infection. While the precise mechanism remains unknown, they suggest that dysregulated activation of innate-like B cells or continued exposure to phosphatidylserine antigens could sustain antibody production in susceptible individuals.
Importantly, the authors note that whether these antibodies actively contribute to disease or simply mark immune dysfunction remains uncertain. “The presence of autoantibodies does not always drive autoimmune disease,” the authors write, emphasizing that further mechanistic studies are needed.
The investigators also draw parallels with syphilis, another spirochete infection. In syphilis, lipid-directed antibody tests are routinely used alongside pathogen-specific assays to monitor disease activity and treatment response. The authors suggest a similar strategy could eventually enhance Lyme disease diagnostics.
“The addition of antilipid antibodies to these panels may improve sensitivity while retaining the more specific diagnostic antibodies,” the authors write. They note that anti-lipid antibodies could potentially function both as adjuncts for early diagnosis and as biomarkers for monitoring recovery following treatment.
The study has several limitations, including relatively modest sample sizes and the use of specimens collected from multiple biobanks with differing collection protocols. Nevertheless, the consistency of findings across independent cohorts strengthens confidence in the observations.
Ultimately, the authors conclude that “the antibodies described here may be valuable biomarkers of early or persistent disease and suggest another mechanism linking B. burgdorferi infection and pathologic autoimmunity.” Larger prospective studies will be needed to determine whether these antibodies can be incorporated into clinical practice for diagnosing Lyme disease or identifying patients at risk for persistent symptoms.
Huntington’s disease therapeutics have reached a historic milestone—the first patient has successfully received an experimental neural stem cell therapy at UCI Health. This groundbreaking dose marks the world’s first human trial of embryonic stem cell-derived neural stem cells for the devastating neurodegenerative disorder.
The treatment, performed in May at University of California Irvine (UCI) Health, represents the culmination of more than 12 years of laboratory research and eight years of clinical planning led by scientists and physicians at the University of California, Irvine. Researchers hope the treatment, known as hNSC-01, could eventually slow disease progression, protect vulnerable brain cells and potentially restore damaged neural circuits.
To date, the first participant has not reported any serious adverse effects, according to the clinical team. A second patient is expected to receive the therapy in July.
Leslie M. Thompson, PhD, Donald Bren Professor of psychiatry and human behavior, as well as neurobiology and behavior, at the University of California, Irvine [UC Irvine]
“This clinical trial highlights the important role that an interdisciplinary academic and clinical team, together with the HD families, plays in advancing medicine,” Leslie M. Thompson, PhD, clinical trial sponsor as well as the Donald Bren Professor of psychiatry and human behavior UC Irvine, told Inside Precision Medicine. “We are grateful to our patients and their incredible families for their bravery to provide hope for others with very few options.”
hNSC-01
Huntington’s disease, caused by a mutation in the huntingtin gene, destroys brain cells, causing involuntary movements, cognitive decline, and psychiatric symptoms that begin between 35 and 50 and worsen over time. Without a cure, the fatal disorder burdens patients and families emotionally, physically, and financially, often requiring daily and long-term care.
Current treatments for Huntington’s disease primarily focus on managing symptoms rather than altering the underlying disease process. Drugs such as tetrabenazine and deutetrabenazine can reduce involuntary movements known as chorea, while antidepressants, antipsychotics and mood stabilizers help address psychiatric symptoms. Physical therapy, speech therapy and occupational therapy can also improve quality of life. However, none of these approaches has been shown to slow or stop the progressive loss of neurons that drives the disease.
Over the past decade, researchers have pursued several experimental disease-modifying strategies. Among the most advanced are gene-targeting therapies designed to reduce production of the mutant huntingtin protein. These include antisense oligonucleotides (ASOs), which are delivered through repeated spinal injections, as well as RNA-targeting and gene-editing approaches intended to suppress or correct the faulty gene. While these strategies directly target the genetic cause of Huntington’s disease, clinical results have been mixed, and questions remain about long-term effectiveness, safety and the need for lifelong treatment.
The hNSC-01 neural stem cell therapy being tested at UCI Health takes a different approach. Rather than targeting the mutant gene itself, the therapy aims to protect vulnerable neurons, replace lost cells, rebuild damaged neural circuits and provide supportive factors that promote brain health.
The UCI researchers believe stem cell-based therapies may offer a new approach by addressing multiple aspects of the disease simultaneously. The experimental treatment, hNSC-01, consists of pluripotent neural stem cells derived from embryonic stem cells and manufactured through the UC Davis Good Manufacturing Practice facility.
Preclinical studies in animal models suggested the cells could perform several functions relevant to Huntington’s disease, including protecting existing neurons, replacing cells that have been lost, rebuilding damaged neural networks and releasing beneficial proteins such as brain-derived neurotrophic factor (BDNF). The cells were also shown to reduce harmful protein accumulations associated with neurodegeneration and demonstrated long-term safety in mice.
Unlike conventional drug therapies, the stem cells are delivered directly into the brain. During the approximately six-hour procedure, performed under general anesthesia, patients lie face down within an MRI scanner while neurosurgeons use a specialized stereotactic navigation and delivery system to implant the cells into the striatum, a deep brain structure heavily affected by Huntington’s disease.
The striatum plays a central role in motor control, decision-making, motivation and learning. Degeneration of this region contributes significantly to the hallmark symptoms of the disorder. The first intervention was delivered by UCI Health neurosurgeon Jefferson W. Chen, MD, and a multidisciplinary surgical team.
Tracking treatment impact
As a Phase Ib/IIa study, the trial’s primary objective is to evaluate safety. However, researchers will also track biomarkers and clinical indicators that may provide early clues about whether the treatment is affecting disease progression.
When asked which biomarkers would help identify how the therapy is working in patients, Thompson emphasized that current measurements are focused more on assessing treatment impact than revealing biological mechanisms. “We will be including HD relevant clinical endpoints and biomarkers, including NfL in plasma and NfL and PENK in CSF; however, these are geared to understanding whether the treatment is having a benefit to these outcome measures versus informing the mechanism of action,” Thompson said.
One of the most important early indicators will be whether disease-related biomarkers remain stable rather than continuing their expected decline. “The earliest sign first and foremost is safety in this initial trial,” Thompson said. “Initial signs that the therapy could be meaningfully altering disease progression would be if the blood-based or CSF-based biomarkers do not show progression.”
Reaching the point of treating the first patient required overcoming a series of scientific, manufacturing and logistical hurdles. According to Thompson, selecting the optimal cell line was among the most significant challenges, testing multiple cell lines in vitro and in vivo.
Researchers also had to establish quality-control standards for the final therapeutic product and create Good Manufacturing Practice cell banks following extensive testing in Huntington’s disease mouse models. The COVID-19 pandemic introduced additional delays. “Disruptions caused by COVID-19, in particular the safety and tumorigenicity studies, delayed the timeline,” Thompson explained.
Another major undertaking involved creating the clinical infrastructure necessary for a first-of-its-kind procedure. Thompson said that it’s not really a challenge, but getting the overall procedural pipeline in place is the first study of this kind at the UCI Health–Irvine hospital in the MRI suite.
Despite the complexity of the project, Thompson said interactions with regulators proceeded smoothly. “We actually had a very good experience in terms of regulatory activities. A very helpful pre-pre-IND, pre-IND and relevant feedback from the FDA on the clinical trial.”
Scalability and competitive landscape
Whether hNSC-01 will ultimately compete with or complement emerging gene-targeting therapies remains unclear. Gene-silencing approaches may be easier to distribute because they do not require brain surgery, but repeated administrations over many years could result in substantial cumulative costs. In contrast, hNSC-01 involves a specialized MRI-guided neurosurgical procedure that may initially be limited to major medical centers, but it is designed as a one-time treatment whose long-term costs could compare favorably with chronic therapies if benefits prove durable.
Thompson believes the infrastructure requirements may be less of a barrier than many assume. “Yes, major medical centers can eventually offer it, and several medical centers are now using this system for other indications,” she said. “The other aspect is this would be a one-time administration so an individual could even travel to a medical center that offers the procedure.”
The REGEN4HD trial plans to enroll 21 adults aged 18 to 65 with early-stage Huntington’s disease. Twelve participants will be included in a Phase Ib dose-escalation cohort, while nine additional participants will be enrolled in a Phase IIa expansion group. The study is funded through a $12 million grant from the California Institute for Regenerative Medicine and coordinated through the UC Irvine Alpha Clinic, one of nine state-supported regenerative medicine clinical research centers.
Even if the therapy proves safe and beneficial, researchers caution that it remains unclear whether stem cell transplantation alone will be sufficient to combat Huntington’s disease over the long term. “At this point we do not know whether this will be sufficient alone or will need to be delivered with other disease-modifying therapies,” said Thompson. “For example, ones that specifically target an HD mechanism such as somatic repeat instability,” Thompson said. “However, these cells also have the potential to exert therapeutic effects directly while serving as vehicles for the delivery of additional interventions.”
For families affected by Huntington’s disease, the first successful treatment in the REGEN4HD trial represents more than a scientific milestone. It marks the beginning of a new chapter in regenerative medicine—one that researchers hope could eventually transform the outlook for a disease that has long remained untreatable.
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.
AI agents are not your “coworkers”
Imagine coming in to work to learn that a new underling will report to you. The worker is not a person but an AI tool—one that your company nonetheless calls Alex, an “employee” with a title and defined responsibilities. How well do you think you would work with Alex?
If you’re anything like the managers studied by Boston University professor Emma Wiles, treating that AI as a “coworker” would lead you to do a worse job. They caught 18% fewer errors when the work was attributed to an agentic “AI employee” rather than a chatbot.
This is an alarming glimpse of the future Silicon Valley is hurling us toward. Microsoft, OpenAI, Anthropic, and Google have all released tools for managing teams of AI agents, many of which are advertised as digital colleagues. Find out why that’s a losing proposition for workers.
—James O’Donnell
This story is from The Algorithm, our weekly AI newsletter. Sign up to receive it in your inbox every Monday.
This flying solar-powered platform could deliver better internet from the air
As soon as August, a giant silver bullet will cut its way through the dry air of the southwestern US and cross the Pacific to reach the coast of Japan.
Once there, the roughly 200-foot-long craft, built by the New Mexico–based company Sceye, will park some 18 kilometers above the ocean’s surface in the stratosphere, then use a custom-built antenna to supplement a 5G network, in a test that includes beaming data straight to devices.
Longevity’s next frontier: “reprogramming” your body
Billions of dollars are flooding into efforts to reverse aging as scientists explore ways to return cells to a younger state. But how far off are these experimental treatments? Will they really work? At a virtual Roundtables event today, MIT Technology Review will examine the science behind the hype.
Science editor Mary Beth Griggs and senior biotechnology reporter Jessica Hamzelou will explore longevity’s latest frontier in a subscriber-only discussion.
I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.
1 The US House has passed new youth online safety legislation + It would set baseline federal standards for kids’ online safety. (Politico $) + States would be allowed to adopt more aggressive protections. (Reuters $) + But critics say it lets tech companies avoid accountability. (Axios) + And tech groups warn it threatens privacy and free expression. (NBC) + The Senate is expected to push for tougher rules. (The Hill)
2 Ford is rehiring human engineers after AI failed to match quality checks It said the AI lacked the training and expertise of technicians. (Bloomberg $) + The new hires will train younger staff and reprogram AI tools. (BBC) + Many firms that replaced workers with AI are now rehiring humans. (Forbes) + The AI jobs hysteria needs a reality check. (MIT Technology Review)
3 Senator Mark Warren is set to introduce a bill to regulate AI agents It would set rules for agent permissions and verification. (The Information $) + Voters of both parties want tighter AI regulation. (NBC News) + But politicians are bitterly divided on the rules. (MIT Technology Review)
4 Rocket Lab is buying Iridium for $8 billion to take on SpaceX It wants to integrate the satellite network with its launch services. (The Verge) + Which could create a fleet that can compete with SpaceX. (WSJ $)
5 Hackers have exposed secrets about Apple’s upcoming iPhone 18 The data was stolen from Tata Electronics, Apple’s Indian supplier. (Reuters $) + The breach also exposed Tesla secrets. (TechCrunch)
6 Chatbots are replacing therapists despite lacking scientific evidence Experts question their safety and therapeutic quality. (WSJ $) + Chatbots may make us lose control of our brains. (MIT Technology Review)
7 Newborn DNA sequencing is edging closer to routine healthcare Trials are expanding despite privacy and ethical concerns. (Economist $) + The push for perfect babies is an ethical mess. (MIT Technology Review)
8 Astronomers are using AI to find new galaxies New tools are reviving decades of space telescope data. (FT $)
9 Remote-controlled cockroach swarms can now breathe underwater The cyborg insects could one day explore Mars. (New Scientist $)
10 Drone shows are creating new forms of worship Churches are depicting biblical stories with thousands of UAVs. (Wired $))
Quote of the day
“This is taking us back to the 1950s, and that is not progress.”
—Edwin Lyman, director of nuclear power safety at the Union of Concerned Scientists, tells NPR that slashing regulations undoes decades of safety lessons from the industry.
One More Thing
Design thinking was supposed to fix the world. Where did it go wrong?
When Kyle Cornforth walked into IDEO’s San Francisco offices for a meeting about reimagining school lunches, she was impressed. “It was Post-its everywhere, prototypes everywhere,” she recalls. “What I really liked was that they offered a framework for collaboration and creation.”
Cornforth was new to IDEO’s way of working: a six-step methodology for innovation called design thinking. But when she looked at the ideas themselves, she had questions: “I was like, ‘You didn’t talk to anyone who works in a school, did you?’ They were not contextualized in the problem at all.”
Design thinking broadened the idea of “design,” elevating designers to take on big, knotty problems through a structured process. But critics argue it has produced unrealistic ideas and, by centering designers, reinforced existing inequities.
A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.)
+ A London tube station has solved its persistent flooding issue by reintroducing beavers. + The Beastie Boys song “Sabotage” has been stunningly recreated in this stop-motion video. + Classical antiquity is lovingly preserved in this collection of over 8,000 late Latin and Greek letters from the Roman world. + This homemade jet-powered fishing boat is a reminder that great engineering and good judgment don’t always travel together.
Top image credit: Photo Illustration by Sarah Rogers/MITTR | Photos Getty
You can follow me on LinkedIn. Thanks for reading!
—Thomas
Posted on
<![CDATA[This episode, titled “Integrating Novel Mechanisms into Schizophrenia Treatment Practice,” features panelist Mark Jankelow discussing the practical clinical considerations surrounding a novel muscarinic agent.
]]>
