Connect with us

Health

Could LLMs Revolutionize Drug and Material Design?

These researchers have developed an innovative system that augments an LLM with graph-based AI models, designed specifically to handle molecular structures

Published

on

Credits:Image: MIT News; iStock

A new method is changing the way we think about molecule design, bringing us closer to the possibility of using large language models (LLMs) to streamline the creation of new medicines and materials. Imagine asking, in plain language, for a molecule with specific properties, and receiving a comprehensive plan on how to synthesize it. This futuristic vision is now within reach, thanks to a collaboration between researchers from MIT and the MIT-IBM Watson AI Lab.

A New era in molecular discovery

Traditionally, discovering the right molecules for medicines and materials has been a slow and resource-intensive process. It often involves the use of vast computational power and months of painstaking work to explore the nearly infinite pool of potential molecular candidates. However, this new method, blending LLMs with other machine-learning models known as graph-based models, offers a promising solution to speed up this process.

These researchers have developed an innovative system that augments an LLM with graph-based AI models, designed specifically to handle molecular structures. The approach allows users to input natural language queries specifying the desired molecular properties, and in return, the system provides not only a molecular design but also a step-by-step synthesis plan.

LLMs and graph models

LLMs like ChatGPT have revolutionized the way we interact with text, but they face challenges when it comes to molecular design. Molecules are graph structures—composed of atoms and bonds—which makes them fundamentally different from text. LLMs typically process text as a sequence of words, but molecules do not follow a linear structure. This discrepancy has made it difficult for LLMs to understand and predict molecular configurations in the same way they handle sentences.

To bridge this gap, MIT’s researchers created Llamole—a system that uses LLMs to interpret user queries and then switches between different graph-based AI modules to generate molecular structures, explain their rationale, and devise a synthesis strategy. The system combines the power of text, graphs, and synthesis steps into a unified workflow.

As a result, this multimodal approach drastically improves performance. Llamole was able to generate molecules that were far better at meeting user specifications and more likely to have a viable synthesis plan, increasing the success rate from 5 percent to 35 percent.

Llamole’s success lies in its unique ability to seamlessly combine language processing with graph-based molecular modeling. For example, if a user requests a molecule with specific traits—such as one that can penetrate the blood-brain barrier and inhibit HIV—the LLM interprets the plain-language request and switches to a graph module to generate the appropriate molecular structure.

This switch occurs through the use of a new type of trigger token, allowing the LLM to activate specific modules as needed. The process unfolds in stages: the LLM first predicts the molecular structure, then uses a graph neural network to encode the structure, and finally, a retrosynthetic module predicts the necessary steps to synthesize the molecule. The seamless flow between these stages ensures that the LLM maintains an understanding of what each module does, further enhancing its predictive accuracy.

“The beauty of this is that everything the LLM generates before activating a particular module gets fed into that module itself. The module is learning to operate in a way that is consistent with what came before,” says Michael Sun, an MIT graduate student and co-author of the study.

Simplicity meets precision

One of the most striking aspects of this new method is its ability to generate simpler, more cost-effective molecular structures. In tests, Llamole outperformed other LLM-based methods and achieved a notable 35 percent success rate in retrosynthetic planning, up from a mere 5 percent with traditional approaches. “On their own, LLMs struggle to figure out how to synthesize molecules because it requires a lot of multistep planning. Our method can generate better molecular structures that are also easier to synthesize,” says Gang Liu, the study’s lead author.

By designing molecules with simpler structures and more accessible building blocks, Llamole could significantly reduce the time and cost involved in developing new compounds.

The road ahead

Though Llamole’s current capabilities are impressive, there is still work to be done. The researchers built two custom datasets to train Llamole, but these datasets focus on only 10 molecular properties. Moving forward, they hope to expand Llamole’s capabilities to design molecules based on a broader range of properties and improve the system’s retrosynthetic planning success rate.

In the long run, the team envisions Llamole serving as a foundation for broader applications beyond molecular design. “Llamole demonstrates the feasibility of using large language models as an interface to complex data beyond textual description, and we anticipate them to be a foundation that interacts with other AI algorithms to solve any graph problems,” says Jie Chen, a senior researcher at MIT-IBM Watson AI Lab.

With further refinements, Llamole could revolutionize fields from pharmaceuticals to material science, offering a glimpse into the future of AI-driven innovation in molecular discovery.

Health

Giant Human Antibody Found to Act Like a Brace Against Bacterial Toxins

This synergistic bracing action gives IgM a unique advantage in neutralizing bacterial toxins that are exposed to mechanical forces inside the body

Published

on

Scientific illustration showing a large human antibody (IgM) attaching to and stabilizing a spiky bacterial toxin protein, visually representing how IgM acts as a physical brace against bacterial toxins to protect human cells
Illustration depicting a giant human antibody (IgM) mechanically bracing a spiky bacterial toxin protein, inspired by recent research on antibodies acting as mechanical stabilizers against bacterial toxins rather than just chemical blockers. Image credit: EdPublica

Our immune system’s largest antibody, IgM, has revealed a hidden superpower — it doesn’t just latch onto harmful microbes, it can also act like a brace, mechanically stabilizing bacterial toxins and stopping them from wreaking havoc inside our bodies.

A team of scientists from the S.N. Bose National Centre for Basic Sciences (SNBNCBS) in Kolkata, India, an autonomous institute under the Department of Science and Technology (DST), made this discovery in a recent study. The team reports that IgM can mechanically stiffen bacterial proteins, preventing them from unfolding or losing shape under physical stress.

“This changes the way we think about antibodies,” the researchers said in a media statement. “Traditionally, antibodies are seen as chemical keys that unlock and disable pathogens. But we show they can also serve as mechanical engineers, altering the physical properties of proteins to protect human cells.”

Unlocking a new antibody role

Our immune system produces many different antibodies, each with a distinct function. IgM, the largest and one of the very first antibodies generated when our body detects an infection, has long been recognized for its front-line defense role. But until now, little was known about its ability to physically stabilize dangerous bacterial proteins.

The SNBNCBS study focused on Protein L, a molecule produced by Finegoldia magna. This bacterium is generally harmless but can become pathogenic in certain situations. Protein L acts as a “superantigen,” binding to parts of antibodies in unusual ways and interfering with immune responses.

Image credit: PIB

Using single-molecule force spectroscopy — a high-precision method that applies minuscule forces to individual molecules — the researchers discovered that when IgM binds Protein L, the bacterial protein becomes more resistant to mechanical stress. In effect, IgM braces the molecule, preventing it from unfolding under physiological forces, such as those exerted by blood flow or immune cell pressure.

Why size matters

The stabilizing effect depended on IgM concentration: more IgM meant stronger resistance. Simulations showed that this is because IgM’s large structure carries multiple binding sites, allowing it to clamp onto Protein L at several locations simultaneously. Smaller antibodies lack this kind of stabilizing network.

“This synergistic bracing action gives IgM a unique advantage in neutralizing bacterial toxins that are exposed to mechanical forces inside the body,” the researchers explained.

The finding highlights an overlooked dimension of how our immune system works — antibodies don’t merely bind chemically but can also act as mechanical modulators, physically disarming toxins.

Such insights could open a new frontier in drug development, where future therapies may involve engineering antibodies to stiffen harmful proteins, effectively locking them in a harmless state.

The study suggests that by harnessing this natural bracing mechanism, scientists may be able to design innovative treatments that go beyond traditional antibody functions.

Continue Reading

Health

How a South African Hospital Team Pioneered the World’s First AI-Powered Cancer Treatment Revolution

Digital Healing: How Bloemfontein Became Ground Zero for the AI Cancer Treatment Revolution

Published

on

Illustrative image for representative purpose/EdPublica

The University of the Free State (UFS), South Africa, and Universitas Academic Hospital have achieved a global healthcare milestone by becoming the first clinical site worldwide to successfully integrate artificial intelligence into cancer treatment planning, marking a transformative advancement in oncology care, according to a statement issued by UFS.

AI implementation

The Departments of Medical Physics and Oncology at UFS, in partnership with Universitas Academic Hospital, have implemented the Radiation Planning Assistant (RPA), a sophisticated web-based AI platform developed by MD Anderson Cancer Center in Houston, Texas. This pioneering initiative has already treated nearly 50 patients, positioning the Bloemfontein-based teams as global leaders in the clinical application of AI in radiotherapy.

Under the leadership of Dr. William Shaw, Senior Lecturer and Deputy Manager in the Department of Medical Physics, the institution has built a robust academic partnership with Professor Laurence Court and his team at MD Anderson Cancer Center—a collaboration that is now yielding remarkable real-world results.

“The introduction and clinical integration of the RPA at the UFS and Universitas Hospital represents a major advancement for oncology services—both regionally and nationally,” Dr. Shaw explained. “It signifies the transition from research collaboration to real-world application, where artificial intelligence is being used to improve access to safe, high-quality cancer care.”

Revolutionizing treatment planning

The RPA technology addresses one of the most time-consuming aspects of cancer care: creating patient-specific radiation treatment plans. The cloud-based platform automates critical components of the treatment planning process, enabling consistent production of high-quality radiotherapy plans while reducing demands on specialized clinical staff.

Dr. Shaw described the streamlined process: “The process begins with the acquisition of a planning CT scan, which serves as the sole imaging input to the RPA. Once the CT dataset has been captured, it is uploaded to the RPA platform via a secure web interface.”

The system uses advanced machine learning algorithms to automatically identify and delineate both tumour volumes and critical normal tissues. Following the completion of the contouring process, the platform automatically generates a comprehensive radiotherapy treatment plan.

Expanding treatment applications

Initially implemented for cervix cancer treatment—representing the largest proportion of radiotherapy patients at the institution—the RPA has since expanded to encompass breast cancer, head and neck cancers, and primary brain tumors. With ongoing institutional support, the system shows significant promise for broader application across nearly all major tumor types treated with external beam radiotherapy.

Professor Vasu Reddy, Deputy Vice-Chancellor for Research and Internationalisation at UFS, praised the achievement: “We extend our congratulations to our colleagues for their exemplary collaborative achievements. Your pioneering work represents the transformative power of multidisciplinary research in advancing medical science and improving patient outcomes.”

Immediate patient benefits

The technology delivers immediate, meaningful improvements for cancer patients by enabling faster access to well-constructed, evidence-based treatment plans reviewed and refined by experts. This translates to more timely care, fewer unplanned treatment interruptions, and improved protection of normal tissues, resulting in fewer side effects and better overall outcomes.

“Our aim is to use artificial intelligence not as a shortcut, but as a tool to standardize, scale, and improve cancer care in places where the need is greatest,” Dr. Shaw emphasized. “The RPA enhances the quality, consistency, and timeliness of cancer treatment in radiotherapy settings—particularly in environments where clinical capacity is limited.”

International expansion

The success in Bloemfontein serves as a model for broader health system innovation, providing a foundation for the safe, phased rollout of similar systems in other provinces. Professor Court has already extended access to the RPA to other radiotherapy centers in South Africa, with expansion to additional countries planned for the near future.

The Department of Oncology, led by Professor Alicia Sherriff, has joined the initiative as an active clinical partner, establishing a multi-disciplinary collaboration that lays the foundation for further research and innovation at the intersection of medical physics, oncology, and data science.

Advanced treatment techniques

Beyond external beam radiotherapy, the UFS and Universitas teams are advancing the use of interstitial brachytherapy for cervix cancer. While not the first globally to implement this specialized technique, the Bloemfontein team ranks among the earliest adopters on the African continent, helping expand access to this advanced modality where it’s most needed.

Future vision

This work received support from the Nuclear Technologies in Medicine and the Biosciences Initiative (NTeMBI), a national technology platform developed and managed by the South African Nuclear Energy Corporation (Necsa) and funded by the Technology Innovation Agency (TIA).

Dr. Shaw’s team has played a central role in developing safe, reliable clinical processes to integrate AI tools like the RPA into daily practice, ensuring that automation enhances rather than replaces professional expertise.

Professor Reddy outlined the broader vision, “The future we are heading towards is one where human innovation and digital technologies work together to elevate the standard of care, rather than replace humanity in medicine. It is encouraging to see how our colleagues are internationalizing our footprint, together with machine precision to enhance detection, personalize treatment and, perhaps importantly, empowering clinicians with data-driven insights for patient care.”

This innovation represents a significant step forward for cancer care in South Africa and demonstrates how international partnerships can bring cutting-edge technologies to healthcare frontlines, making them work effectively in real clinics for real patients. As cancer incidence rises across low- and middle-income countries, the leadership shown by the UFS and Universitas teams offers a compelling model for how academic medical centers can respond with agility, scientific rigor, and global solidarity.

Edited by Chris Jose

Continue Reading

Health

Researchers Develop Low-Cost Sensor for Real-Time Detection of Toxic Sulfur Dioxide Gas

Sulfur dioxide, a toxic air pollutant primarily released from vehicle exhaust and industrial processes, is notorious for triggering respiratory irritation, asthma attacks, and long-term lung damage.

Published

on

In a significant breakthrough for environmental monitoring and public health, scientists from the Centre for Nano and Soft Matter Sciences (CeNS), Bengaluru, India, have developed an affordable and highly sensitive sensor capable of detecting sulfur dioxide (SO₂) gas at extremely low concentrations.

Sulfur dioxide, a toxic air pollutant primarily released from vehicle exhaust and industrial processes, is notorious for triggering respiratory irritation, asthma attacks, and long-term lung damage. Monitoring its presence in real time is essential, but existing technologies are often expensive, power-hungry, or ineffective at detecting the gas at trace levels.

To address this gap, the CeNS team, under the leadership of Dr. S. Angappane, has engineered a novel sensor by combining two metal oxides — nickel oxide (NiO) and neodymium nickelate (NdNiO₃). NiO serves as the receptor that captures SO₂ molecules, while NdNiO₃ acts as a transducer that converts the chemical interaction into an electrical signal. This innovative design enables the sensor to detect SO₂ at concentrations as low as 320 parts per billion (ppb), outperforming many commercial alternatives.

Speaking about the development, Dr. Angappane said in a media statement, “This sensor system not only advances the sensitivity benchmark but also brings real-time gas monitoring within reach for a wider range of users. It demonstrates how smart materials can provide practical solutions for real-world environmental challenges.”

Threshold-triggered sensor response in a) Safe state, b) Warning state, and c) Danger state. Image credit: PIB

The CeNS team has also built a portable prototype incorporating the sensor. It features a user-friendly threshold-triggered alert system with color-coded indicators: green for safe levels, yellow for warning, and red for danger. This visual approach ensures that even non-specialist users can understand and respond to pollution risks instantly. Its compact size and lightweight design make it ideal for deployment in industrial zones, urban neighborhoods, and enclosed environments requiring continuous air quality surveillance.

The sensor system was conceptualized and designed by Mr. Vishnu G Nath, with key contributions from Dr. Shalini Tomar, Mr. Nikhil N. Rao, Dr. Muhammed Safeer Naduvil Kovilakath, Dr. Neena S. John, Dr. Satadeep Bhattacharjee, and Prof. Seung-Cheol Lee. The research findings were recently published in the journal Small.

With this innovation, CeNS reinforces the role of advanced materials science in developing cost-effective technologies that protect both public health and the environment.

Continue Reading

Trending