Why do some people breeze through viral infections which leave others fighting for their lives? Could we find special molecules that coax cellular proteins to work in our favour, disarming harmful proteins or helping protective ones to stick around? And in a world overflowing with data, could light-powered computers, modelled after the human brain, revolutionise how we process information?
These questions drive three unique research journeys led by rising scientists at A*STAR. This year, they were selected for the prestigious 2025 National Research Foundation (NRF) Fellowships, which recognise Singapore’s most promising early-career researchers and their visionary, high-impact projects in science and engineering.
Open to scientists of all nationalities and disciplines, the NRF Fellowship (NRFF) provides a competitive, five-year research grant to chart bold directions in science and technology.
In this issue of A*STAR Research, we sit down with the 2025 cohort of NRF Fellows to uncover what inspires their curiosity, the big questions they are determined to answer, and how their discoveries might shape our world in the years to come.
Yi-Hao Chan
Principal Investigator
Genetics of Host Immunity Lab
A*STAR Infectious Diseases Labs (A*STAR IDL)
Tell us about the viral diseases you’ve studied.
I’ve had the opportunity to work on a range of viral diseases that are important both in Singapore and globally.
During my PhD degree, I studied the disease mechanics of several mosquito- or arthropod-borne viruses (arboviruses) such as chikungunya (CHIKV), Zika and O’nyong‘nyong (ONNV). One of our key findings was that CD4+ T cells—immune cells that coordinate our response to infection—contribute to joint inflammation in patients with CHIKV or ONNV. This discovery opens new avenues for drug development.
When the COVID-19 pandemic hit, I seized the chance to contribute to Singapore’s national health response by working closely with clinicians in Singapore’s public hospitals. We found that T cell responses, in particular T Helper 2 responses, were important in controlling SARS-CoV-2’s disease severity. This was evident when we compared patients who remained asymptomatic to those who developed COVID-19 pneumonia. These insights were especially valuable during the early days of the pandemic when much about COVID-19 was still unknown.
For my second round of postdoctoral training at the Rockefeller University, US, under Jean-Laurent Casanova’s tutelage, I studied why certain individuals develop severe brain inflammation after infection by viruses like herpes simplex virus (HSV) and SARS-CoV-2. Though common, these viruses can cause devastating outcomes in rare cases. We found that these can be due to single-gene mutations that affect the body’s antiviral defences. Identifying these genetic vulnerabilities not only reveals essential parts of our immune systems, but also uncovers potential targets for future treatments against infectious diseases.
What links our genes and viral disease outcomes?
While studying genetic factors in HSV infection, I identified its first human restriction factor in the brain: a transmembrane protein called TMEFF1. Restriction factors are natural proteins that block viruses from entering and replicating in cells, acting as part of the body’s built-in antiviral defence.
In two unrelated cases of herpes simplex encephalitis (HSE)—a rare but serious brain infection—we discovered harmful mutations in the TMEFF1 gene that either disabled the protein or misdirected it within the cell. This resulted in the virus gaining unrestricted entry into brain cells and ultimately triggering severe inflammation in the brain.
This discovery has important biological and medical implications, as there is currently no vaccine for HSV-1, and its standard treatment—acyclovir—only targets general viral replication. A TMEFF1-based therapy that blocks viral entry, whether as a soluble protein or as an antibody-drug conjugate, would be a new and more targeted way to treat not just HSE, but also common conditions of mucosal and skin surfaces, such as cold sores. We are now working to translate this discovery into clinical use.
What will the NRFF award help you achieve?
The NRFF is a major boost to my goal of tackling the infection enigma around viral infections: that is, why the same virus can lead to vastly different clinical outcomes across infected individuals. Why do some patients experience severe illness while others don’t? By easing our lab’s mental bandwidth and resource constraints, the award lets us explore more ideas and make more discoveries in this area.
Our research is focused on uncovering the genetic and immunological factors behind life-threatening viral diseases. We are recruiting patients who developed severe illness alongside individuals who remain asymptomatic or have only mild symptoms despite exposure to the same virus. By genetically sequencing both groups, we aim to pinpoint key genetic variants that cause life-threatening disease outcomes. We also aim to identify unique, essential antiviral pathways that dampen disease, which would in turn deepen our understanding of human biology.
What’s next for your lab?
I plan to grow my research group over the next year and delve deeper into deadly arboviruses such as dengue. In 2024 alone, there were more than 13,000 reported cases of dengue in Singapore, and in some years—such as 2020—that number climbed as high as 35,000. Rising global temperatures are expanding mosquito populations, which are in turn driving arbovirus transmission and escalating the threat they pose to public health.
Since my work centres on human genetics, it needs access to cohorts of patients with infectious diseases. I’m actively seeking partnerships with infectious disease clinicians who are keen to unravel the infection enigma and to advance more personalised, mechanism-targeting treatments for affected patients.
What would you say to younger aspiring scientists?
Persevere, even when the going gets tough. Keep pushing forward and surround yourself with people who uplift and challenge you. I highly encourage finding an experienced mentor who is willing to guide and support you. I’ve personally benefited from mentors who believed in me, especially during moments of doubt.
Don’t give up!
Shuang Liu
Principal Investigator
A*STAR Institute of Molecular and Cell Biology (A*STAR IMCB)
Take us down the journey in science that led you here.
My passion lies in transforming how we approach drug discovery, a field at the intersection of many scientific disciplines. I began my journey in medicinal chemistry as an undergraduate at Imperial College London, UK, where I was supported by A*STAR’s National Science Scholarship and studied how small molecule drugs are designed and made.
At A*STAR’s Experimental Therapeutic Centre (ETC), now the Experimental Drug Development Centre (EDDC), I had my first experience in fragment-based screening for promising drug candidates. For my PhD degree at the University of Oxford, UK, I investigated how drugs interact with proteins, with a focus on isocitrate dehydrogenase (IDH)—a key enzyme mutated in brain cancer.
As a postdoctoral fellow at the Broad Institute, US, I pioneered a targeted way to discover molecular glue-like compounds using DNA-encoded library (DEL) screening: a method that uses DNA barcode tags on millions to billions of small molecules to rapidly screen for protein binding. Molecular glues had traditionally been found by serendipity, but we were among the first to systematically identify them using a high-throughput platform.
Returning to A*STAR IMCB, I was honoured to receive the A*STAR Young Achiever Award, which supported the launch of my independent research in this area. Each phase of my training—from building molecules, to understanding how they work in cells, and to innovating methods to screen them—has shaped the integrated research approach that was recognised by the NRFF award. I’ve come to appreciate that meaningful breakthroughs often emerge not from complexity, but from the elegant combination of simple ideas across disciplines.
How can molecular glues help treat cancer and other difficult diseases?
Molecular glues are small molecules that work differently from most traditional drugs. Rather than targeting a single protein, they bring together two proteins that don’t normally interact—typically a disease-related target and a ‘presenter’ that influences the target’s fate.
This mechanism is powerful because it’s versatile enough to modulate a wide range of cellular processes. For instance, if the presenter is an E3 ubiquitin ligase—a protein that tags others for destruction—molecular glues can drive the breakdown of harmful proteins, such as those linked to cancer and neurodegenerative diseases. Conversely, when paired with a presenter such as deubiquitinase, which removes those destruction tags, glues can also stabilise beneficial proteins like tumour suppressors.
What makes molecular glues especially exciting is their ability to target proteins previously considered ‘undruggable’, which make up over 80 percent of the human proteome—the full set of proteins our cells can make. This offers enormous potential for treating complex diseases.
What does the NRFF award enable you to work on next?
The award gives me the resources and collaborative environment to build on the momentum of my work and take it to the next level. It allows me to assemble a team of talented, motivated scientists who share the same vision, and marks a pivotal step in my longer journey to push the boundaries of chemical biology and drug discovery.
At the heart of my project is the goal of accelerating the discovery of molecular glues using DEL screening. While DEL is already transforming drug discovery, it hasn’t been widely applied to the search for new monovalent molecular glues. With the NRFF’s support, I can develop a robust screening platform for that very purpose.
As a proof-of-concept, I will apply this platform in two biological systems—first, the induced breakdown, via molecular glues, of drug-resistant, cancer-causing protein mutants. The next would go beyond degradation: the disruption of harmful protein complexes, using molecular glues that can pull key proteins out of those complexes.
Insights from these systematic screens will also guide the design of entirely new families of molecular glues. Ultimately, this could transform how we tackle hard-to-treat diseases.
What do you hope to achieve in the coming years?
I aim to conduct a series of proof-of-concept studies showing how different presenter proteins can regulate specific cellular functions. While most research on molecular glues has focused on protein degradation, I’m particularly keen on exploring their other effects. These include disrupting protein-protein interactions, stabilising useful proteins, and modulating cell signalling pathways. These studies will not only advance fundamental science but also provide the blueprints for new therapeutic strategies.
On the translational side, I plan to develop promising molecules identified through DEL screening into preclinical candidates for cancer treatment, working closely with drug development experts at EDDC. The focus on oncology is just the start; the same approach could be extended to many other disease areas.
To bring this vision to life, I look to collaborate with academic groups, biotech firms and pharmaceutical companies interested in applying molecular glue strategies to their own targets of interest. I also warmly welcome young scientists eager to contribute to and grow with this emerging field.
What advice would you offer aspiring early-career researchers?
Identify what makes your research and training unique. Carve out a niche where your skills and experiences give you an edge in tackling important scientific questions. Becoming a top expert in your field can set you apart. For me, a deep understanding of IDH’s biochemical mechanisms ultimately informed DEL screening strategies that others might not have considered for different protein targets.
Don’t shy away from ambitious goals. Sometimes, it is the seemingly modest pursuits that lead to the most transformative insights. Be patient and diligent in building your track record. The dots often only connect in hindsight, but the depth and integrity of your work will speak for itself.
Stay curious, stay committed, and the rest will follow.
Bowei Dong
Principal Investigator
A*STAR Institute of Microelectronics (A*STAR IME)
What excites you about your recent work?
I study emerging high performance computing systems known as photonic neuromorphic computing (PNC), which process information using the dynamics of light and draw inspiration from the brain’s architecture. These systems can be designed to handle the ultrafast data transfer and energy efficiency demands of modern artificial intelligence (AI) models. However, they typically need high-quality light sources to maintain stable optical signals—sources that are challenging to produce and operate.
In our recent work, we demonstrated something surprising: PNC can actually work better using lower-quality light sources, if you design a system right. These light sources are much easier to manufacture, cost less, consume less energy, and are simpler to control—yet they can boost the system’s computing performance.
When we used this technology to identify patients with Parkinson’s disease, we achieved an accuracy of 92 percent. This discovery pushes the boundaries of what was previously thought possible and opens new doors to more affordable, energy-efficient AI technologies. Our findings have been published in Nature and featured in the journal’s “Seven Technologies to Watch in 2025” list.
In your opinion, how close is PNC to real-world use?
We can be very optimistic about PNC’s adoption in ways that make real-world impact. In the international sphere, two leading startups have recently unveiled PNC prototypes.
There’s Lightmatter, which reported developing a universal photonic AI processor that can run versatile advanced AI models and deep reinforcement learning algorithms, which rely on a lot of trial and error to optimise decisions. As their processor delivers near-electronic precision on many tasks, it’s a notable entry for photonic computing to compete with established electronic AI accelerators.
The other startup, Lightelligence, has reported an integrated large-scale photonic accelerator with over 16,000 light-based components, achieving ultralow response times of just three nanoseconds. It can handle the same computational workload as a commercial high-performance A10 graphics processor—but using light instead of electricity as a data medium.
Beyond these design and prototype breakthroughs, we’re also seeing momentum on the mass manufacturing side. Taiwan Semiconductor Manufacturing Company (TSMC)—the world's leading chipmaker that produces semiconductors for other companies—has begun investing in photonic computing. That’s a strong sign that the industry sees it as the next near-term tech wave.
Some PNC systems are already at mid-to-high technology readiness levels. At this pace, it is reasonable to expect PNC architectures in real-world settings within the next two to five years.
What does the NRFF award mean to you?
The award is a meaningful recognition of the potential of PNC and my work in this field. It reflects the confidence of the research community in PNC as they place their bets on its development. It also highlights NRF and A*STAR’s commitment to investing in innovation and deep tech by supporting early-career researchers such as myself in tackling some of the world’s most challenging but critical problems.
This prestigious award also affirms my dedication to research and strengthens my resolve to push the boundaries of PNC. It motivates me to climb R&D’s steepest mountains with greater confidence.
With the NRF’s generous support, I aim to develop the world’s best PNC system: a light-based computer so powerful, it can handle massive amounts of data at unprecedented speeds with extreme energy efficiency. The key innovation involved is a photonic computing system that sees the big picture and functions as a unified whole, rather than a collection of smaller parts.
This goes beyond faster internet searches or smoother movie streaming; we’re aiming for a future where computers can run self-driving cars, real-time speech recognition and other complex AI applications in the blink of an eye, using less power than current technologies.
What research goals lie ahead for you?
My long-term goal is to push for the market adoption of PNC technology and establish it as a core component of future computing systems—much like the CPU is today. Achieving this vision will require building a strong global ecosystem through close collaboration between industry, research and technology organisations (RTOs) and institutes of higher learning (IHLs).
In the near term, with the support of the NRFF award and A*STAR, I aspire to build up the PNC capabilities at A*STAR IME, which provides world-class photonics R&D facilities and the multidisciplinary expertise essential for developing practical PNC systems.
I also plan to collaborate with other RTOs and IHLs to address key PNC challenges—such as scalability—and nurture a new generation of R&D talent and leadership to propel the field forward. Furthermore, by working with public agencies, I hope to expand industry outreach efforts to anchor PNC R&D activities in Singapore.
These combined efforts will drive a vibrant PNC R&D ecosystem and bring us closer to taking this technology to market.
What is your advice for younger researchers?
Take action now—there’s no better time than the present. And remember, failure isn’t a setback, but a valuable learning experience.
