Workshop on “Research on AChE inhibitory compounds for the treatment of dementia using a combination of machine learning (ML) modeling, molecular dynamics simulation and in vitro testing”

To facilitate interaction, exchange, and sharing of research findings between the Institute for Advanced Study of Technology (IAST) and the Faculty of Pharmacy (Ton Duc Thang University), alongside external collaborative partners such as: the Institute of Natural Products Chemistry (INPC, Vietnam Academy of Science and Technology, Hanoi), and Hanoi University of Pharmacy, with the aim of strengthening scientific research collaboration, the Institute for Advanced Study of Technology is organizing a specialized seminar on "Investigating AChE Inhibitory Compounds for Alzheimer's Disease Treatment Using an Integrated Approach of Machine Learning (ML) Modeling, Molecular Dynamics Simulations, and In Vitro Experiments" with the following details:

Details:

• Time: 8:30 AM, May 20, 2025
• Location: IAST Office in Hanoi
• Format: Hybrid (In-person and Online)

The seminar will feature presentations by six speakers, including:

1. Assoc. Prof. Dr. Pham Minh Quan: "Research on Screening for Hepatoprotective Compounds from Vietnamese Natural Active Substances"

• Abstract: Liver cancer, with hepatocellular carcinoma (HCC) being the most prevalent type in humans, ranks as the second leading cause of cancer-related deaths worldwide. Current treatment methods remain largely ineffective, driving efforts to discover novel molecules or therapeutic strategies against HCC. Mortalin, an hsp70 chaperone protein, is overexpressed in various cancers, including HCC. Mortalin binds to p53, thereby preventing p53's nuclear translocation and inhibiting its tumor-suppressing functions. Consequently, inhibiting mortalin–p53 interactions has been proposed as an anticancer drug development strategy. In silico screening of a natural compound database identified solasonine, a steroidal glycoalkaloid from the Solanaceae family, as a potent inhibitor of p53–mortalin interactions. Pharmacological studies confirmed that solasonine can effectively suppress mortalin p53 interaction in the HepG2 HCC cell line expressing both mortalin and p53, leading to p53 nuclear translocation.

2. Assoc. Prof. Dr. Tran Phuong Thao: "Research and Development of Novel Compounds for Alzheimer's Disease Treatment: In Silico and Experimental Approaches"

• Abstract: Alzheimer's disease accounts for approximately 60% to 80% of all conditions causing memory, cognitive, and behavioral decline in humans. Alzheimer's disease has created a significant disease and economic burden on healthcare systems, with Alzheimer's patients requiring increasing levels of medical care as the disease progresses. Research and development of novel compounds for Alzheimer's disease treatment using a combination of in silico and experimental methods is a promising direction that has attracted the attention of scientists.

3. MSc. Do Anh Tuan: "Accurate structure prediction of biomolecular interactions with AlphaFold 3"

• Abstract: The introduction of AlphaFold 2 has spurred a revolution in modelling the structure of proteins and their interactions, enabling a huge range of applications in protein modelling and design. Here we describe our AlphaFold 3 model with a substantially updated diffusion-based architecture that is capable of predicting the joint structure of complexes including proteins, nucleic acids, small molecules, ions and modified residues. The new AlphaFold model demonstrates substantially improved accuracy over many previous specialized tools: far greater accuracy for protein–ligand interactions compared with state-of-the-art docking tools, much higher accuracy for protein–nucleic acid interactions compared with nucleic-acid-specific predictors and substantially higher antibody–antigen prediction accuracy compared with AlphaFold-Multimer v.2.3. Together, these results show that high-accuracy modelling across biomolecular space is possible within a single unified deep-learning framework.

4. Pharmacist Thai Quynh Mai: "Identification of AChE targeted therapeutic compounds for Alzheimer’s disease: an in-silico study with DFT integration"

• Abstract: Alzheimer’s disease (AD) is a progressive neurodegenerative condition marked by cognitive deterioration and changes in behavior. Acetylcholinesterase (AChE), which hydrolyzes acetylcholine, is a key drug target for treating AD. This research aimed to identify new AChE inhibitors using the IMPPAT database. We used known drugs as a basis to search for similar chemicals in the IMPPAT database and created a library of 127 plant-based compounds. Initial screening of these compounds was performed using molecular docking, followed by an analysis of their drug-likeness and ADMET properties. Compounds with favorable properties underwent density functional theory (DFT) calculations to assess their electronic properties such as HOMO-LUMO gap, electron density, and molecular orbital distribution. These descriptors provided insights into each compound’s reactivity, stability, and binding potential with AChE. Promising candidates were further evaluated through molecular dynamics (MD) simulations over 100 ns and MMPBSA analysis for the last 30 ns. Two compounds, Biflavanone (IMPHY013027) with a binding free energy of − 130.394 kcal/mol and Calomelanol J (IMPHY007737) with − 107.908 kcal/mol, demonstrated strong binding affinities compared to the reference molecule HOR, which has a binding free energy of − 105.132 kcal/mol. These compounds exhibited promising drug-ability profiles in both molecular docking and MD simulations, indicating their potential as novel AChE inhibitors for AD treatment. However, further experimental validation is necessary to verify their effectiveness and safety.

5. Engineer Nguyen Thi Nguyet Hang: "Molecular docking as a tool for the discovery of molecular targets of nutraceuticals in diseases management"

• Abstract: Molecular docking is a computational technique that predicts the binding affinity of ligands to receptor proteins. Although it has potential uses in nutraceutical research, it has developed into a formidable tool for drug development. Bioactive substances called nutraceuticals are present in food sources and can be used in the management of diseases. Finding their molecular targets can help in the creation of disease-specific new therapies. The purpose of this review was to explore molecular docking’s application to the study of dietary supplements and disease management. First, an overview of the fundamentals of molecular docking and the various software tools available for docking was presented. The limitations and difficulties of using molecular docking in nutraceutical research are also covered, including the reliability of scoring functions and the requirement for experimental validation. Additionally, there was a focus on the identification of molecular targets for nutraceuticals in numerous disease models, including those for sickle cell disease, cancer, cardiovascular, gut, reproductive, and neurodegenerative disorders. We further highlighted biochemistry pathways and models from recent studies that have revealed molecular mechanisms to pinpoint new nutraceuticals’ effects on disease pathogenesis. It is convincingly true that molecular docking is a useful tool for identifying the molecular targets of nutraceuticals in the management of diseases. It may offer information about how nutraceuticals work and support the creation of new therapeutics. Therefore, molecular docking has a bright future in nutraceutical research and has a lot of potentials to lead to the creation of brand-new medicines for the treatment of disease.

6. Engineer Bui Nguyen Thanh Long: "Peptide-binding specificity prediction using fine-tuned protein structure prediction networks"

• Abstract: Peptide-binding proteins play key roles in biology, and predicting their binding specificity is a long-standing challenge. While considerable protein structural information is available, the most successful current methods use sequence information alone, in part because it has been a challenge to model the subtle structural changes accompanying sequence substitutions. Protein structure prediction networks such as AlphaFold model sequence structure relationships very accurately, and we reasoned that if it were possible to specifically train such networks on binding data, more generalizable models could be created. We show that placing a classifier on top of the AlphaFold network and fine-tuning the combined network parameters for both classification and structure prediction accuracy leads to a model with strong generalizable performance on a wide range of Class I and Class II peptide-MHC interactions that approaches the overall performance of the state-of the-art NetMHCpan sequence-based method. The peptide-MHC optimized model shows excellent performance in distinguishing binding and non-binding peptides to SH3 and PDZ domains. This ability to generalize well beyond the training set far exceeds that of sequence-only models and should be particularly powerful for systems where less experimental data are available.