Department of Integrative Biotechnology, College of Biotechnology and Bioengineering, Sungkyunkwan University, Suwon, 16419, Gyeonggi-do, Republic of Korea
Auto-generated profile, based on publicly available information
Duong Thanh Tran
Department of Integrative Biotechnology, College of Biotechnology and Bioengineering, Sungkyunkwan University, Suwon, 16419, Gyeonggi-do, Republic of Korea
Peptide-based therapeutics hold great promise for the treatment of various diseases; however, their clinical application is often hindered by toxicity challenges. The accurate prediction of peptide toxicity is crucial for designing safe peptide-based therapeutics. While traditional experimental approaches are time-consuming and expensive, computational methods have emerged as viable alternatives, including similarity-based and machine learning (ML)-/deep learning (DL)-based methods. However, existing methods often struggle with robustness and generalizability. To address these challenges, we propose HyPepTox-Fuse, a novel framework that fuses protein language model (PLM)-based embeddings with conventional descriptors. HyPepTox-Fuse integrates ensemble PLM-based embeddings to achieve richer peptide representations by leveraging a cross-modal multi-head attention mechanism and Transformer architecture. A robust feature ranking and selection pipeline further refines conventional descriptors, thus enhancing prediction performance. Our framework outperforms state-of-the-art methods in cross-validation and independent evaluations, offering a scalable and reliable tool for peptide toxicity prediction. Moreover, we conducted a case study to validate the robustness and generalizability of HyPepTox-Fuse, highlighting its effectiveness in enhancing model performance. Furthermore, the HyPepTox-Fuse server is freely accessible at https://balalab-skku.org/HyPepTox-Fuse/ and the source code is publicly available at https://github.com/cbbl-skku-org/HyPepTox-Fuse/. The study thus presents an intuitive platform for predicting peptide toxicity and supports reproducibility through openly available datasets.
26 September 2024 in Expert Systems with Applications
Large language models (LLMs) have significantly advanced computational biology by enabling the integration of molecular, protein, and natural language data to accelerate drug discovery. However, existing molecular captioning approaches often underutilize diverse molecular modalities and lack interpretability. In this study, we introduce XMolCap, a novel explainable molecular captioning framework that integrates molecular images, SMILES strings, and graph-based structures through a stacked multimodal fusion mechanism. The framework is built upon a BioT5-based encoder-decoder architecture, which serves as the backbone for extracting feature representations from SELFIES. By leveraging specialized models such as SwinOCSR, SciBERT, and GIN-MoMu, XMolCap effectively captures complementary information from each modality. Our model not only achieves state-of-the-art performance on two benchmark datasets (L+M-24 and ChEBI-20), outperforming several strong baselines, but also provides detailed, functional group-aware, and property-specific explanations through graph-based interpretation. XMolCap is publicly available at https://github.com/cbbl-skku-org/XMolCap/ for reproducibility and local deployment. We believe it holds strong potential for clinical and pharmaceutical applications by generating accurate, interpretable molecular descriptions that deepen our understanding of molecular properties and interactions.
Despite the scale-up of vector control, diagnosis and treatment, and health information campaigns, malaria persists in the forested areas of South-Central Vietnam, home to ethnic minority populations. A mixed-methods study using an exploratory sequential design was conducted in 10 Ra-glai villages in Bac Ai district of Ninh Thuan province to examine which social factors limited the effectiveness of the national malaria elimination strategy in the local setting. Territorial arrangements and mobility were found to directly limit the effectiveness of indoor residual spraying and long-lasting insectidical treated nets (LLINs). Households (n=410) were resettled in the “new villages” by the government, where they received brick houses (87.1%) and sufficient LLINs (97.3%). However, 97.6% of households went back to their “old villages” to continue slash-and-burn agriculture. In the old village, 48.5% of households lived in open-structured plot huts and only 5.7% of them had sufficient LLIN coverage. Household representatives believed malaria could be cured with antimalarials (57.8%), but also perceived non-malarial medicines, rituals, and vitamin supplements to be effective against malaria. Household members (n = 1,957) used public health services for their most recent illness (62.9%), but also reported to buy low-cost medicines from the private sector to treat fevers and discomfort as these were perceived to be the most cost-effective treatment option for slash-and-burn farmers. The study shows the relevance of understanding social factors to improve the uptake of public health interventions and calls for contextually adapted strategies for malaria elimination in ethnic minority populations in Vietnam and similar settings.
In the field of image and video colorization, the existing research employs a CNN to extract information from each video frame. However, due to the local nature of a kernel, it is challenging for CNN to capture the relationships between each pixel and others in an image, leading to inaccurate colorization. To solve this issue, we introduce an end-to-end network called Vitexco for colorizing videos. Vitexco utilizes the power of the Vision Transformer (ViT) to capture the relationships among all pixels in a frame with each other, providing a more effective method for colorizing video frames. We evaluate our approach on DAVIS datasets and demonstrate that it outperforms the state-of-the-art methods regarding color accuracy and visual quality. Our findings suggest that using a ViT can significantly enhance the performance of video colorization.
GinDB is a comprehensive database containing 753 standardized ginsenosides with curated and corrected data from 1963 to 2024. Integrated with GinDB-AI, a deep learning module for predicting bioactivity and physicochemical properties, the platform enables rapid structure-activity analysis and visualization. The freely accessible resource accelerates ginsenoside-based drug discovery and natural product research.