Scientific Programs

Prof. Chi-Huey Wong

Decoding Protein Glycosylation for Vaccine and Antibody Development

Glycosylation is an important biological process for modulating the structure and function of proteins, cells, and many other biologics. However, this process has not been well understood due to the lack of tools and methods available for the study of biological glycosylation. Decoding glycan structures and protein glycosylation will thus help us understand the role of post-translational glycosylation with molecular precision and provide new opportunities for developing better glycoprotein medicines to ameliorate diseases associated with aberrant glycosylation. Over the years, we have been actively involved in developing new tools and methods, notably the chemoenzymatic and programmable one-pot methods, for making and studying complex glycans and glycoproteins, and investigating the impact of glycosylation on protein folding, viral infection, cancer progression, and immune responses. This lecture highlights the advanced glycosylation methods developed in our laboratory that have been used to drive new discoveries in glycobiology and accelerate the translation of new discoveries to innovative developments. Representative examples include practical and expedient synthesis of oligosaccharides and glycoproteins, development of glycan microarrays, low-sugar universal vaccines against human viruses with deletion of host-made glycan shields to elicit broadly protective immune responses, and cell-based methods with humanized glycosylation pathway to produce monoclonal antibodies with Fc-glycosylation optimized for antibody-mediated target killing. It is hoped that advances in glycosylation methodology and the extensive data generated over the years by the glycoscience community, combined with AI assistance, will lead to a paradigm change in vaccine and antibody development as well as drug discoveries for human health. 

Prof. Dean Ho

Biohacking Actual: Precision Made Personal

In 2024, Prof. Dean Ho and team launched DELTA, a first-in-kind interventional human trial – with Prof. Dean Ho as the test subject. This N=1 protocol harnesses a combination of AI, digital medicine, fasting, fitness, and food to fortify metabolic health, monitored using an array of digital health platforms. Built from an unprecedented dataset, this study will culminate in a digital twin of Dean to hyper-personalise his cardiometabolic health protocol. Outcomes from this trial will also power population-scale healthspan optimisation regimens.

Prof. Sanghamitra Bandyopadhyay

Machine Learning Applications in Healthcare

Machine Learning (ML) algorithms find extensive applications in all domains of scientific research, in particular in diverse areas of biology and healthcare for making novel discoveries and for gaining deeper insights into various processes of life. In this talk we will first present brief overviews of artificial intelligence, machine learning and molecular biology. In particular, a quick introduction to the central dogma of molecular biology will be provided, which is fundamental to the understanding of a large class of machine learning applications in biology. Various kinds of data sets emerging in different areas of healthcare research will be mentioned. We will then discuss some case studies in a few areas of biology, namely, molecular target prediction, a graph theoretic method for biomarker identification and graph based approaches in drug interaction studies. The talk will conclude with a mention of some issues and challenges in this area. 

Prof. Thomayant Prueksaritanont

Roles and Applications of Predictive ADMET in Drug Discovery and Development: Chula4DR Program Examples

Predictive ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) modeling has become a fundamental component of modern drug discovery and development, providing a framework for the early assessment of pharmacokinetic and toxicological characteristics prior to experimental validation. By integrating advanced computational tools and machine learning–based algorithms, predictive ADMET enables effective compound prioritization, optimizes lead selection, and significantly reduces the number and cost of downstream clinical trials. Three case studies from Chula4DR—spanning drug discovery, development, and clinical pharmacology—will be presented to illustrate its effectiveness in minimizing resource-intensive experimental workflows and enhancing the translational value of preclinical research. Collectively these examples underscore the essential role of predictive ADMET in advancing rational drug design and fostering a more efficient, knowledge-based pharmaceutical development paradigm.

Dr. Jung-Hsin Lin

Generative AI and Robotics for Biomedical Applications and Drug Discovery

In this speech,  I will highlight some major research topics of my lab on the developments and applications of computational methodologies for design and discovery of new drugs, and for unraveling the molecular mechanisms of biological systems based on fundamental physical chemical principles, which are facilitated by biophysical experiments, as well as molecular modelling and simulations. I will also describe an integrated virtual screening scheme established in our lab, which combines rapid docking and deep-learning-based scoring approaches, molecular dynamics simulations and adaptive umbrella sampling methods, enabling the screening of very large libraries of chemical compounds in a very efficient manner while preserving high accuracy.  I will also give some examples that are quite challenging for AI-based structure prediction, as well as our recently established automatic laboratory systems with well-controlled experimental environment and AI robotics for carrying out cell culture. virus titration assays, chemical synthesis, and drug testing.

Porf. KWOH CHEE KEONG

Converging AI, Sensing, and Bioinformatics for Precision Healthcare and Drug Innovation

The convergence of Artificial Intelligence (AI), bioinformatics, and sensing technologies is fundamentally reshaping the discovery and delivery of healthcare. This keynote traces a research journey that mirrors this evolution—from developing foundational computational models to realizing their translational impact in precision medicine and drug innovation.

We begin with the core computational challenges of modern bioinformatics: predicting protein–protein interactions, detecting protein complexes, and integrating heterogeneous biological networks to elucidate disease mechanisms. These foundational methods underpin the rational design of therapeutics, extending into machine learning models for drug–target interaction prediction, binding affinity estimation, and covalent docking, thereby accelerating the identification and optimization of novel drug candidates.

A complementary research direction on decoding genomic and molecular determinants of disease and drug response. Integrative AI approaches are applied to identify disease genes, infer synthetic lethality in cancer, model signaling pathway dynamics, and track viral evolution—from influenza to SARS-CoV-2—to inform vaccine design and antigenic variant prediction.

Together, these developments illustrate how converging AI, sensing, and bioinformatics can form a unified computational ecosystem that transforms multi-scale biomedical data into actionable intelligence, driving the next generation of precision healthcare and drug innovation.

Prof. Maria Siluvay Michael Gromiha

Machine Learning and AI Based Methods for Identifying Potential Inhibitors and Disease Causing Mutations in Proteins: Applications to Drug Design

The substitution of amino acid residues in a protein alters its structure, stability and function, and may lead to diseases. We have developed comprehensive databases for annotating disease-causing mutations in globular and membrane proteins using experimental data reported in the literature. These databases provide a description of subcellular location of the mutant, structural environment and functional features along with visualization, search, and display, and download options. We have systematically analyzed the effect of these mutations at different levels such as (i) specific targets (ii) protein classes including membrane proteins and (iii) different cancer types, and different perspectives: (i) preference of residues at mutant positions, (ii) probability of substitutions, (iii) influence of neighboring residues in driver and passenger mutations and (iv) distribution of driver and passenger mutations around hotspot sites.  Further, we have developed machine learning, large language model and AI based methods for predicting disease causing and cancer-specific mutations at a large scale. On the other hand, we have developed a pipeline for identifying potential inhibitors for targets, which are involved in cancer. These methods have been utilized for identifying the potential driver and passenger mutations at a large scale, as well as lead compounds, which could be used for designing experiments. The salient features of the results will be discussed.

Prof. Lee-Wei Yang

Gene-Specific Pathogenicity Prediction DNN Outperforms AlphaMissense and Predict the Inheritance Mode of Hearing Loss Variants by First Principles

The integration of artificial intelligence with molecular simulations is transforming the landscape of predictive genomics and precision drug discovery.  While deep multimodal models such as AlphaMissense have advanced the prediction of amino acid variant pathogenicity, their dependence on weakly labeled evolutionary data and massive parameter counts limits interpretability, updatability, and sustainability.  Variants of uncertain significance (VUS) continue to far exceed experimentally validated cases, highlighting the need for frameworks that couple explainable AI with physical realism.

Prof. Chandra S. Verma

Computational Discovery Guiding Clinical Decisions

Computational models of diverse types including sequence based and structure based have been informing biomolecular mechanism, drug discovery very effectively. Given the current nature of sophistication in methods and in computer hardware, we will discuss how, with an iterative team effort, the state of the art can actually assist in clinical decision making, most notably has been successful in repurposing.

Prof. Suree Jianmongkol

Bridging Traditions, Cultivating the Future: Agro-pharmacy as Thailand's Path to Sustainable Healthcare

The global and Thai healthcare system face mounting pressure from aging populations, rising chronic disease and medical expense, with dependency on imported pharmaceuticals. Thailand, now an aged society, urgently needs sustainable solutions, especially with a projected shortage of 10,000 pharmacists by 2034. This necessitates an innovative approach: integrating Thai Traditional Medicine (TTM) with modern pharmaceutical science.  This integration promises enhanced prevention, reduced import dependency, novel drug discovery avenues from rich biodiversity, and holistic patient care, supporting national goals like the BCG Economy and Medical Hub status.

Central to this integration is Agro-pharmacy, a visionary concept championed by the new Faculty of Pharmaceutical Sciences at Kasetsart University (KU). Agro-pharmacy is the synergistic application of agricultural and pharmaceutical sciences to develop high-quality medicinal products from sustainable agricultural and herbal resources. This approach covers the entire value chain, from cultivation to clinical application, moving beyond the traditional dichotomy of pharmaceutical sciences (industrial focus) and pharmaceutical care (clinical focus). Its aim is to cultivate a new generation of pharmacists capable of leveraging Thailand's unique agricultural strengths for self-reliance and sustainable development.

Kasetsart University (KU), as Thailand's premier agricultural institution, is uniquely positioned to lead this initiative through its integrated ecosystem, including the KU Medical Park and specialized research centers. The new Doctor of Pharmacy (Pharm.D.) program, launching in 2026, embodies this vision. The curriculum integrates Agro-pharmacy principles alongside core pharmaceutical competencies, with dedicated courses like "Introduction to Agro Pharmacy" and "Agro-Based Product Development and Innovation" series.  This program aims to equip pharmacists with unique skills, such as producing and analyzing Agro-pharmacy products and developing innovative formulations from agricultural outputs.

In summary, Agro-pharmacy offers a transformative bridge between Thailand's rich traditions and modern scientific potential, paving the way for a more sustainable, self-reliant, and effective healthcare system. By cultivating pharmacists skilled in this integrated approach—possessing expertise beyond the traditional scope, particularly in leveraging agricultural resources for health innovation—Kasetsart University is seeding the future of Thai healthcare. Realizing this vision requires robust collaboration among academia, industry, policymakers, and communities to create lasting opportunities for health and national development.

Prof. Ivy Chung

Exploring Drug Repurposing for Equitable Healthcare in Global South

  The rising burden of non-communicable diseases, coupled with the high cost and long timelines of new drug development, underscores the urgent need for affordable therapeutic strategies in the Global South. Drug repurposing—the systematic identification of new indications for existing or discontinued drugs—offers a pragmatic and cost-effective pathway to expand treatment access while reducing R&D risks. This approach can accelerate translation, particularly in regions where access to novel therapeutics remains limited.

  Within the framework of the Universiti Malaya Affordable Diagnostics and Therapeutics (UMADT) program, under the International Affordable Diagnostics & Therapeutics Alliance (IA-DATA), several initiatives exemplify this vision. UMADT integrates bioinformatics, clinical data mining, and mechanistic studies to identify candidate drugs for cancer and infectious diseases, emphasizing relevance to regional disease patterns. Collaborative networks link academic, clinical, and policy stakeholders to advance preclinical validation, translational research, and equitable deployment.

  By leveraging existing pharmacological assets and regional expertise, drug repurposing provides a bridge between scientific innovation and public health need. Advancing this paradigm through South–South collaboration can democratize drug discovery, strengthen local innovation ecosystems, and ensure that life-saving therapies reach all populations—safely, rapidly, and affordably.

Prof. Partha Roy

Kaempferol: A major phytochemical present in various natural products plays critical role in protecting pancreatic β-cells and preventing obesity related diabetes

Pancreatic β-cells are affected by fatty acids which plays a vital role in the pathological manifestation of obesity linked to type II diabetes. Thus, rescuing β-cells from fatty acid induced apoptosis is linked to prevent obesity related type II diabetes. Kaemferol, a natural flavonoid present in honey, has been previously shown to have extensive therapeutic implications for its inherent anti-oxidative, anti-inflammatory, anticancer and anti-microbial activities. In the present study, we intended to determine the cytoprotective effect of kaemferol on pancreatic β-cells undergoing apoptosis under the palmitic acid-stressed condition. We found that kaemferol could show prominent increase in cell viability by attenuating palmitic acid-induced lipotoxicity of pancreatic β-cells. The protective effect by kaempferol was through inhibition of apoptosis and up-regulation of autophagy. The study was confirmed by both in vitro and in vivo analysis. Our data showed that kaemferol also up and down-regulates phosphorylation of AMPK and mTOR respectively. Subsequently, upon inhibition of AMPK phosphorylation by compound C (an inhibitor of AMPK), kaemferol mediated autophagy was abolished which further led to the decline in β-cell survival. Such observations collectively lead to the conclusion that, kaemferol exerts its cytoprotective role against lipotoxicity by activation of autophagy via AMPK/mTOR pathway.    

Prof. Yi-Cheng Chang

Preclinical Drug Development for Obesity, Diabetes, and Kidney Disease

Peroxisome proliferator-activated receptor γ (PPARγ) is a master transcriptional regulator of systemic insulin sensitivity and energy balance. The anti-diabetic drug thiazolidinediones (TZDs) are potent synthetic PPARγ ligands with undesirable side effects, including obesity, fluid retention, and osteoporosis. 15-keto-PGE2 is an endogenous PPARγ ligand metabolized by prostaglandin reductase 2 (PTGR2). Here, we confirmed that 15-keto-PGE2 binds and activates PPARγ via covalent binding. In patients with type 2 diabetes and obese mice, serum 15-keto-PGE2 levels were decreased. Administration of 15-keto-PGE2 improves glucose homeostasis and prevented diet-induced obesity in mice. Either genetic inhibition of PTGR2 or PTGR2 inhibitor BPRPT0425 protected mice from diet-induced obesity, insulin resistance, and hepatic steatosis without fluid retention and osteoporosis. These data indicate inhibition of PTGR2 is a new therapeutic approach to treat diabetes and obesity through increasing endogenous PPARγ ligands without side effects of synthetic PPARγ ligands TZDs.

ALDH2 (acetaldehyde dehydrogenase 2, mitochondrial) is the key metabolizing enzyme of acetaldehyde and other toxic aldehydes, such as 4-hydroxynonenal. A missense Glu504Lys mutation of the ALDH2 gene is prevalent in 560 million East Asians, resulting in reduced ALDH2 enzymatic activity. We find that Aldh2 knock-in mice mimicking human Glu504Lys mutation were prone to develop diet-induced obesity, glucose intolerance, insulin resistance, and fatty liver due to reduced adaptive thermogenesis and energy expenditure. AD-9308, a water-soluble, potent, and highly selective ALDH2 activator ameliorates diet-induced obesity and fatty liver, and improves glucose homeostasis in both Aldh2 wild-type and knock-in mice. These results highlight the potential of reducing toxic aldehyde levels by activating ALDH2 for treating obesity.

Approximately 400 million people worldwide carry genetic variants in glucose-6-phosphate dehydrogenase (G6PD), making G6PD mutation the second most common monogenic mutation globally. In Taiwan, it is estimated that 1.6% of the population (2.8% of males and 0.7% of females) carry G6PD gene mutations, with the c.1376G>T (p.R459L) point mutation, also known as the Canton type favism variant, being the most common. We have preliminarily observed that knock-in mice mimicking the Canton-type favism mutation develop chronic kidney injury. High-through output drug screening identified several hits for G6PD activator.

We also identified a glucose polymer can effectively absorb intestinal fat and reduce diet-induced obesity by approximately ~10% in high-fat high-sucrose diet-induced obese rats. This fat sequestrant can complement the shortcomings of current incretin-based therapy.

Dr. Shu-Chin Su Chen

A Medical Device Company’s AI Journey

As a medical device company, our AI journey focuses on using machine learning to enhance device performance, improve patient outcomes, and drive innovation while maintaining regulatory compliance.

Prof. Yang-Chang Wu

The Value-Added of Heat-Clearing and Detoxifying Traditional Chinese Medicine

Traditional Chinese Medicine (TCM) is well-known for its therapeutic efficacy, bioactive components, and relatively low toxicity. These characteristics make TCM a promising complementary treatment for cancer. Our findings indicate that treatment with heat-clearing and detoxifying TCM formula is associated with the reduced mortality in breast cancer patients. To further evaluate the clinical potential of this TCM formula for cancer therapy, the anti-metastatic effects of two principal heat-clearing and detoxifying herbs were investigated in triple-negative breast cancer (TNBC) models. Water extracts of both herbs individually inhibited TNBC cell migration and invasion without inducing cytotoxicity. When combined, the extracts exhibited a strong synergistic effect, with a combination index of 0.2. In vivo, oral administration markedly suppressed metastasis in 4T1-Luc orthotopic mouse models. Mechanistically, the bioactive compounds coptisine and resveratrol side were identified as key contributors that downregulated the expression of epithelial–mesenchymal transition (EMT)-related markers. Ingenuity Pathway Analysis (IPA) further suggested the involvement of HRH3 and PPM1L signaling pathways. These findings highlight the potential synergistic benefit of combining multiple heat-clearing and detoxifying TCM herbs or their active components as an adjuvant therapy for metastatic TNBC.

Multiple Functions and Benefits