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
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.
COP30
Health Systems ‘Unprepared’ as Climate Impacts Intensify, Experts Warn at COP30
India will require $643 billion between now and 2030 to adapt to climate change under a business-as-usual scenario
On Health Day at COP30 (November 13), global health and climate experts warned that the world is dangerously underprepared for the accelerating health impacts of climate change, calling for a dramatic scale-up of adaptation finance to protect vulnerable populations.
Speaking at a press conference hosted by Regions4, the Global Climate & Health Alliance and CarbonCopy, leaders from research institutions and national governments said climate-linked health threats — from extreme heat to wildfire smoke — are rising sharply while funding remains “colossally” insufficient.
“Each year, more than half a million lives are lost due to heat, and over 150,000 deaths are linked to wildfire smoke exposure,” said Dr. Marina Romanello, Executive Director of the Lancet Countdown. “Health systems, already stretched and underfunded, are struggling to cope with these growing pressures, and most are still unprepared for what is coming.”
Romanello added that despite the scale of the crisis, “only 44% of countries have costed their health adaptation needs, and existing finance falls short by billions. Without urgent investment, we will not be able to protect populations from escalating climate impacts.”
Adaptation gap continues to widen
The speakers described health-sector underfunding as a critical part of the broader adaptation finance gap. The latest UNEP Adaptation Gap Report estimates developing countries will need $310–365 billion annually by 2035, while the international community is still struggling to mobilise even the $40 billion Glasgow Pact Goal.
“With regards to finance, the reality is that we have a deficit that is quite colossal,” said Carlos Lopes, Special Envoy for Africa, COP30 Presidency. “Most of the efforts are from national authorities. What we need from international finance is that it must be complementary.”
Lopes cautioned that climate and health policy still operates in “multiple contested layers,” warning that unless these are aligned, “we risk losing coherence in our global response.”
Countries highlight urgent needs
Representatives from Bangladesh, Nigeria, India and Chile echoed concerns that adaptation finance is far from matching on-ground needs.
“Our adaptation financing for health is far below what is needed. The gap between what we require and what we receive is enormous,” said Md Ziaul Haque, Additional Director General, Ministry of Environment, Bangladesh. He urged multilateral finance entities to bring forward “concrete, holistic proposals that match the scale of the challenge.”
Nigeria’s challenges are equally stark. “In Nigeria, we are facing an additional 21% disease burden due to climate change… yet the adaptation finance we received in 2021–22 only met 6% of our needs,” said Oden Ewa, Commissioner for Special Duties and Green Economy Lead. He called adaptation finance a “lifeline that saves lives, strengthens communities, and protects economies.”
India outlines its adaptation burden
India also presented updated estimates of its climate adaptation needs. “India will require $643 billion between now and 2030 to adapt to climate change under a business-as-usual scenario,” said Dr. Vishwas Chitale of the Council for Energy, Environment & Water. He noted that India has already made “significant progress, spending $146 billion in 2021–2022 alone — 5.6% of GDP.”
New funding coalition signals momentum
Speakers highlighted the launch of the Climate and Health Funders Coalition, which has committed an initial $300 million annually.
“This is an encouraging signal… It shows that the world is beginning to recognise that protecting health must be at the centre of climate adaptation,” said Jeni Miller, Executive Director, Global Climate & Health Alliance.
Health at the centre of adaptation
Chile stressed the need for integrated policy approaches.
“It is vital to combine the efforts of different ministries — not only health but also transport, energy and food production — so that we generate co-benefits across sectors,” said Dr. Sandra Cortes, President of Chile’s Climate Change Scientific Committee. “A more integrated approach will allow us to improve public health, reduce emissions and create fairer, more sustainable development opportunities.”
As negotiators continue discussions in Belém, experts reiterated that adaptation finance — especially for health — must be just, equitable, accessible and prioritise the most climate-vulnerable nations. The recently proposed Belém Health Action Plan and the Global Goal on Adaptation are expected to serve as frameworks for strengthening health system resilience worldwide.
Health
Air Pollution Claimed 1.7 Million Indian Lives and 9.5% of GDP, Finds The Lancet
In 2022 alone, fine particulate pollution — PM2.5 — killed an estimated 1.7 million people in India, according to the Lancet Countdown on Health and Climate Change 2025.
In 2022 alone, fine particulate pollution — PM2.5 — killed an estimated 1.7 million people in India, according to the Lancet Countdown on Health and Climate Change 2025.
The same toxic particles that fill Delhi’s winter air and blanket cities from Kanpur to Kolkata also caused economic losses equivalent to 9.5% of India’s GDP, revealing that air pollution is not just a public health emergency, but a national economic crisis hiding in plain sight.
A Crisis Woven into Everyday Life
India’s worsening air quality is no longer a seasonal problem. According to The Lancet Countdown, over a third of Indians were exposed to PM2.5 levels exceeding World Health Organization (WHO) limits for more than 10 months of the year.
Rising temperatures, urban sprawl, and fossil fuel combustion — from coal-fired power plants to vehicle emissions — have created a deadly feedback loop that is choking the country’s lungs and its economy.
“Air pollution in India is a silent pandemic. It’s not only shortening lives, but undermining productivity, healthcare systems, and economic growth,” said Dr. Marina Romanello, Executive Director of The Lancet Countdown, in the report’s global launch statement.
The Health Toll: From Newborns to the Elderly
The Lancet Countdown 2025 estimates that the global death toll from air pollution reached 8.3 million in 2022, with India accounting for over one-fifth of those fatalities.
PM2.5 — particles less than 2.5 microns in diameter — penetrate deep into lungs and bloodstreams, causing or worsening heart disease, stroke, diabetes, and respiratory illness.
In India, the burden falls disproportionately on the poorest households, who are more likely to live near highways, coal plants, or industrial clusters and have limited access to healthcare.
Children and elderly people are the most vulnerable: the report highlights that exposure to dirty air increases the risk of low birth weight, premature births, and chronic illness later in life.
Counting the Cost: 9.5% of GDP Lost
The Lancet Countdown’s economic assessment, based on lost labour productivity, healthcare costs, and premature deaths, found that India lost 9.5% of its GDP in 2022 due to air pollution-related impacts.
That’s roughly equivalent to USD 300 billion — more than India’s entire annual education and health budgets combined.
Urban centres such as Delhi, Lucknow, and Patna rank among the most polluted in the world.
Air pollution is estimated to reduce life expectancy in northern India by up to 7 years, according to the Energy Policy Institute at the University of Chicago, underscoring how pervasive the damage has become.
“For a fast-growing economy like India, this is a double blow,” said Prof. Randeep Guleria, pulmonologist and former AIIMS director. “It burdens healthcare systems while reducing worker output — exactly the opposite of what a young nation needs.”
Climate and Air: The Same Enemy
The report connects India’s pollution crisis to its dependence on fossil fuels — especially coal — which remains the largest source of both CO₂ and PM2.5 emissions.
While government programmes such as the National Clean Air Programme (NCAP) and electric mobility initiatives aim to reduce pollution, progress has been slow.
Many of the dirtiest thermal plants continue to operate without meeting emission standards, and vehicle emissions remain poorly regulated outside major cities.
“Air pollution is not a separate problem from climate change — it’s the same story told through different symptoms,” noted Dr. Romanello. “Every tonne of coal burned harms both lungs and the climate.”
This linkage is echoed in India’s own National Electricity Plan 2032, which outlines aggressive renewable targets, and in Ember’s 2025 analysis, which found that expanding coal capacity further would be economically irrational — a finding that strengthens the case for rapid decarbonisation.
Health as an Economic Argument
The Lancet Countdown reframes pollution not just as an environmental or health challenge, but as an economic imperative.
In India, labour losses due to heat and pollution exposure have grown by 42% since the early 2000s, with outdoor and informal workers suffering the most.
As heatwaves and smog increasingly overlap, lost work hours and rising healthcare costs could slow GDP growth by up to 1.8 percentage points annually by the mid-2030s if left unchecked.
Experts say cleaner power and transport sectors could deliver rapid wins:
- Phasing out coal and shifting to renewables can cut PM2.5 emissions by over 60% in key industrial zones.
- Expanding public transit and EV adoption can reduce vehicular PM2.5 by one-third in metropolitan regions.
- Strengthening NCAP’s monitoring and enforcement could save hundreds of thousands of lives each year.
From Policy to Breathable Air
Despite India’s national clean air mission and renewable push, enforcement and coordination remain major gaps.
The report calls for integrating air quality and climate policies, arguing that cutting fossil fuel use provides a “double dividend” — cleaner air and fewer greenhouse gases.
This integration has begun in limited form: several Indian states, including Gujarat and Maharashtra, have introduced emissions trading schemes for industrial pollutants.
But experts say scaling such initiatives nationally, alongside stricter vehicle standards and urban planning reforms, is critical for measurable results.
A Moment of Reckoning
The Lancet Countdown 2025 warns that air pollution and climate impacts are already reversing health gains made over decades.
India’s choice is no longer between growth and clean air — it’s about whether growth can continue at all under the weight of rising illness, lost labour, and degraded ecosystems.
“Air pollution is robbing India of its demographic dividend,” the report concludes. “Clean air is not a luxury; it’s a prerequisite for sustainable development.”
As the smog season begins once again in northern India, the data are unambiguous:
The invisible killer is now visible — and unaffordable.
References:
The Lancet Countdown on Health and Climate Change 2025; The Lancet; Energy Policy Institute at the University of Chicago (EPIC); Ember; CREA.
Health
Study Unveils Mucus Molecules That Block Salmonella and Prevent Diarrhea
A new MIT-led study reveals how key mucus molecules naturally shield the gut from dangerous bacteria like Salmonella
A new MIT-led study reveals how key mucus molecules naturally shield the gut from dangerous bacteria like Salmonella. The breakthrough opens new pathways for affordable, preventative treatments for travelers and soldiers at risk of infection.
Researchers at MIT have discovered new powerful ways the body guards itself from dangerous bacteria: by deploying mucins, special molecules in mucus that neutralize microbes and stop infection before it starts. The team identified mucins—especially MUC2 and MUC5AC—in the digestive tract that shut down the genetic machinery Salmonella uses to invade cells and cause diarrhea.
“By using and reformatting this motif from the natural innate immune system, we hope to develop strategies to preventing diarrhea before it even starts. This approach could provide a low-cost solution to a major global health challenge that costs billions annually in lost productivity, health care expenses, and human suffering,” said Katharina Ribbeck, the Andrew and Erna Viterbi Professor of Biological Engineering at MIT, in a media statement.
In experiments, exposing Salmonella to the intestinal mucin MUC2 blocked the bacterial proteins that enable infection, turning off the critical regulator HilD. The study also found that a similar mucin from the stomach, MUC5AC, works the same way—and both molecules appear to protect against multiple foodborne germs triggered by similar genetic switches.
“We discovered that these mucins not only create a physical shield but also actively control whether pathogens can turn on genes needed for infection,” Ribbeck explained in the media statement.
Lead authors Kelsey Wheeler and Michaela Gold say synthetic versions of these mucins could soon be added to oral rehydration salts or chewable tablets, providing practical protection for troops, travelers, and people in high-risk areas. According to Wheeler, “Mucin mimics would particularly shine as preventatives, because that’s how the body evolved mucus — as part of this innate immune system to prevent infection,” she said.
-
Society1 week agoThe Ten-Rupee Doctor Who Sparked a Health Revolution in Kerala’s Tribal Highlands
-
Space & Physics6 months agoNew double-slit experiment proves Einstein’s predictions were off the mark
-
Earth2 months agoData Becomes the New Oil: IEA Says AI Boom Driving Global Power Demand
-
COP302 months agoCorporate Capture: Fossil Fuel Lobbyists at COP30 Hit Record High, Outnumbering Delegates from Climate-Vulnerable Nations
-
COP302 months agoBrazil Cuts Emissions by 17% in 2024—Biggest Drop in 16 Years, Yet Paris Target Out of Reach
-
Women In Science3 months agoThe Data Don’t Lie: Women Are Still Missing from Science — But Why?
-
Space & Physics1 month agoIndian Physicists Win 2025 ICTP Prize for Breakthroughs in Quantum Many-Body Physics
-
Space & Physics6 months agoJoint NASA-ISRO radar satellite is the most powerful built to date


