Huynh Anh Duy

Huynh Anh Duy

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
HDHuynh Anh Duy
Huynh Anh Duy
Tarapong SrisongkramTarapong Srisongkram
Tarapong Srisongkram
Dengue virus (DENV) remains a critical global health challenge, with no approved antiviral treatments currently available. The growing prevalence of DENV infections highlights the urgent need for effective therapeutics. Antiviral peptides (AVPs) have gained significant attention due to their potential to inhibit viral replication. However, traditional drug discovery methods are often time-consuming and resource-intensive. Advances in artificial intelligence, particularly deep generative models (DGMs), offer a promising approach to accelerating AVP discovery. This report provides a comprehensive assessment of the role of DGMs in identifying novel AVPs for DENV. It presents an extensive survey of existing antimicrobial and AVP datasets, peptide sequence feature representations, and the integration of DGMs into computational peptide design. Additionally, in vitro and in silico screening data from previous studies highlight the therapeutic potential of AVPs against DENV. Variational autoencoders and generative adversarial networks have been extensively documented in the literature for their applications in AVP generation. These models have demonstrated a remarkable capacity to generate diverse and structurally viable compounds, significantly expanding the repertoire of potential antiviral candidates. Additionally, this report assesses both the strengths and limitations of DGMs, providing valuable insights for guiding future research directions. As a data-driven and scalable framework, DGMs offer a promising avenue for the rational design of potent AVPs targeting DENV and other emerging viral pathogens, contributing to the advancement of next-generation therapeutic strategies.

21 September 2026 in ChemMedChem
Verified
Journal Article
Open access
PDF file
HDHuynh Anh Duy
Huynh Anh Duy
Tarapong SrisongkramTarapong Srisongkram
Tarapong Srisongkram
Identifying safe and effective anti‐inflammatory agents is challenging, especially given the adverse effects of long‐term conventional therapies. Here, we present a multistage computational framework to prioritize early natural compounds with potential anti‐inflammatory activity across diverse chemical spaces. The proposed workflow integrates machine learning, consensus prediction, drug‐likeness evaluation, synthetic accessibility assessment, toxicity prediction, and molecular docking to enable stepwise refinement of candidate selection. Six predictive models were developed and evaluated with gold‐standard metrics, all showing consistent performance with accuracies, AUROCs, and AUPRCs above 0.8. A consensus strategy was used to reduce model variability and improve prediction reliability. SHAP analysis identified structural scaffolds associated with predicted anti‐inflammatory activity, providing interpretability for compound prioritization. Sequential filtering selected compounds that met predefined criteria for drug‐likeness, synthetic feasibility, and predicted safety profiles. Application of the framework to the NPASS database yielded 20 prioritized natural products, of which 17 candidates were further selected based on molecular docking analysis, exhibiting predicted binding affinities of <−7 kcal/mol. This framework was deployed on a web server, CAIP ( https://caip‐predictor.streamlit.app/ ), enabling rapid prediction and preliminary virtual screening of novel compounds. In total, the proposed framework offers a computational approach to systematically support the prioritization of anti‐inflammatory candidate molecules.

20 September 2026 in ChemRxiv
Verified
Preprint
Open access
HDHuynh Anh Duy
Huynh Anh Duy
SKSastiya Kampaengsri
Sastiya Kampaengsri
SPSupreeya Paiboon
Supreeya Paiboon
Tarapong SrisongkramTarapong Srisongkram
Tarapong Srisongkram
Stevens-Johnson syndrome (SJS) is a rare, life-threatening adverse drug reaction for which early identification of high-risk drugs remains challenging. In this study, we developed an explainable quantitative structure-toxicity relationship (QSTR) framework for SJS prediction using a multi-view learning framework with 21 molecular representations, including molecular descriptors, fingerprints, transformer-based embeddings, and a molecular similarity representation. Model development and evaluation were designed in accordance with the OECD principles for QSAR validation, incorporating scaffold-aware validation, external evaluation on pharmacovigilance negative-signal compounds, and structural interpretation by integrating SHAP with Bemis-Murcko scaffold enrichment analysis. The proposed framework achieved robust sensitivity and specificity, with molecular similarity consistently improving the identification of SJS-associated compounds. Furthermore, a consensus prediction and scaffold-based prioritization strategy identified several high-risk drugs from the 1 external negative-signal dataset, including cefprozil, tetracycline, and omadacycline, highlighting their potential association with SJS and warranting further pharmacovigilance and experimental investigation. To facilitate practical application, this was deployed as a freely accessible web server, SimSJSAlert (https: //sim-sjs-alert.streamlit.app/), enabling rapid drug-induced SJS risk assessment. The proposed framework combines predictive performance, interpretability, and accessibility to support early SJS risk assessment, pharmacovigilance signal prioritization, and safer drug development.

21 September 2026 in ChemRxiv
Verified
Preprint
Open access
Tarapong SrisongkramTarapong Srisongkram
Tarapong Srisongkram
SPSupreeya Paiboon
Supreeya Paiboon
HDHuynh Anh Duy
Huynh Anh Duy
Natural products represent a major source of therapeutic molecules, yet their structural diversity and limited biological annotations pose substantial challenges for computational exploration and discovery. Here, we develop a multimodal molecular representation framework that integrates chemical language, molecular fingerprints, physicochemical descriptors, and molecular graphs into a unified latent space for natural product discovery. Using structurally out-of-distribution molecular spaces, we show that molecular representation and sampling strategy jointly determine which regions of chemical space are explored. Multimodal pretraining captured chemically informative representations that enabled exploration of structurally unseen natural products, while comparisons with singlemodality representations showed that representation advantages depended on the discovery objective rather than being universally superior. The pretrained representation was transferable across anti-inflammatory and antioxidant activity predictions, and its integration with active learning enabled data-efficient exploration of biologically active molecules. Prospective evaluation of structurally unseen natural products further connected model-guided exploration with experimental antioxidant activity. These findings establish molecular representation and sampling as coupled determinants of chemical-space exploration and show

6 June 2024 in TNU Journal of Science and Technology
Verified
Journal Article
Closed access
PDF file
HDHuynh Anh Duy
Huynh Anh Duy
TNTrần Thị Kim Ngà
Trần Thị Kim Ngà
HVHuỳnh Phú Vinh
Huỳnh Phú Vinh
Ludwigia hyssopifolia được sử dụng trong dân gian để điều trị bệnh vàng da ở nhiều nước châu Á. Nghiên cứu nhằm đánh giá đặc tính kháng oxy hóa và ức chế dòng tế bào HepG2 của cao chiết. Cao ethanol và phân đoạn chloroform, ethyl acetat và methanol được chuẩn bị. Mẫu được khảo sát đặc tính kháng oxy hóa trên mô hình DPPH. Thử nghiệm MTT được thực hiện để nghiên cứu hoạt tính ức chế HepG2. Các hợp chất dễ bay hơi từ cao ethanol được xác định bằng GC-MS. Hàm lượng polyphenol và flavonoid toàn phần được xác định bằng phương pháp quang phổ. Cao ethanol ghi nhận 15 cấu tử, với crocetane, heneicosane và 3,5-di-tert-butylphenol là các thành phần chính. Phân đoạn ethyl acetat và methanol có hàm lượng polyphenol và flavonoid cao nhất cũng như có hoạt tính kháng oxy hóa tốt nhất. Cao ethanol toàn phần và phân đoạn chloroform cho hoạt tính ức chế HepG2 tiềm năng nhất. Do đó, L. hyssopifolia có tiềm năng để phát triển các sản phẩm chống oxy hóa và bảo vệ gan.

20 October 2023 in Tạp chí Y Dược học Cần Thơ
Verified
Journal Article
Closed access
PDF file
DVDang Thuan Vo
Dang Thuan Vo
HDHuynh Anh Duy
Huynh Anh Duy
Background: Medicinal plants play an important role in the alternative or complement therapy to manage of urinary stones at this time. Terminalia genus was proved anti-urolithiatic activity via in vitro inhibition of calcium oxalate formation. Among the samples, Terminalia catappa showed as a potential plants for this activity in India. Moreover, they were a common plant species in Vietnam and there was no research on this topic in our country. Objectives: To evaluate in vitro antiurolithiatic activity of Terminalia catappa leaves in Vietnam, through inhibition of calcium oxalate formation. In addition, moisture value, preliminary screening of the chemical composition and determination of tanninoid content of aqueous extract also were conducted. Materials and methods: Moisture content of herbs was conducted according to guidelines of 5th Vietnam pharmacopoeia, appendix 9.6. Qualitation of the phytochemical constituents of aqueous extract with appropriate reagents to confirm the presence of natural compounds via chemical reactions. Determination of tanninoid content of aqueous extract was evaluated by using oxidation-reduction titration method (Lowenthal assay) and skin powder method in instructing of appendix 12.6 from 5th Vietnam pharmacopoeia. Inhibitory effect of calcium oxalate formation of three hydrophilic extracts (45% ethanol, 96% ethanol and aqueous extracts) was confirmed by nucleation assay with cystone as a positive control. Results: Moisture content of medicinal plant was 11.625%. Terminalia catappa leaves contain the major phytochemical constituents such as flavonoids, tanninoids and saponins. The tanninoid content according to Lowenthal method and skin powder method were 10.88% and 10.70%, respectively. Therefore, the average tannin content was confirmed to be approximately 10.79%. Among the investigated samples, aqueous extract showed the best inhibitory activity of calcium oxalate crystal formation with an IC50 of  602.67 μg/mL when compared to cystone with IC50 423.05 μg/mL. Conclusions: The aqueous extract from Terminalia catappa leaves has been shown to be a promising source for anti-calcium oxalate crystals formation activity on experimental model. The anti-urolithic potential of Terminalia catappa leaves may be related to its major phytoconstituents.