個人簡歷

Education/Training:

• 2010, PhD, Institute of Bioinformatics, National Chiao Tung University
• 2005, B.S., Department of Biology, National Cheng Kung University

Professional and Research Experience:

• 2020/08 - 2021/07, Associate Dean, College of Management, TMU
• 2020/08 - 2021/07, Director, International Ph.D. Program in Biotech and Healthcare Management, TMU
• 2019/03 - 2021/07, Associate Professor & Professor, Graduate Institute of Data Science, TMU
• 2011/08 - 2019/03, Assistant Professor & Associate Professor, School of Pharmacy, KMU
• 2012/03 - 2012/05, Visiting scholar, Center for Bioinformatics, NCTR, US FDA
• 2008/09 - 2009/08, Visiting scholar, Institute for Computer Science, University of Tübingen, Germany

Awards and Honors:

• 2023 - Outstanding Alumni (by Department of Life Sciences, National Cheng Kung University)
• 2021 - Excellent Research Paper Award (by Taipei Medical University)
• 2018 - Publons Peer Review Awards 2018 (Top 1% in Multidisciplinary)

Selected Publications:

• Lin RH, Lin P, Wang CC, Tung CW*. A Novel Multitask Learning Algorithm for Tasks with Distinct Chemical Space: Zebrafish Toxicity Prediction as an Example. J Cheminform. 2024, 16(1), 91.
• Wang SS, Wang CC, Wang CL, Lin YC, Tung CW*. Incorporating Tissue-Specific Gene Expression Data to Improve Chemical–Disease Inference of in Silico Toxicogenomics Methods. J Xenobiot. 2024, 14(3), 1023-1035.
• Huang WC, Lin WT, Hung MS, Lee JC, Tung CW*. Decrypting orphan GPCR drug discovery via multitask learning. J Cheminform. 2024, 16(1), 10.
• Chiu YW, Tung CW*, Wang CC*. Multitask learning for predicting pulmonary absorption of chemicals. Food Chem Toxicol. 2024, 185, 114453.
• Wang SS, Lin P, Wang CC, Lin YC, Tung CW*. Machine Learning for Predicting Chemical Migration from Food Packaging Materials to Foods. Food Chem Toxicol. 2023, 178, 113942.

Abstract

Chinese Title:

建構虛擬斑馬魚系統以研析化合物毒性

English Title:

Developing a Virtual Zebrafish System for Studying Chemical Toxicity

Abstract

Among new approach methodologies, computational toxicology models can provide fast and economic evaluation of toxicity with increasing regulatory acceptance. Despite of the extensive use of quantitative structure-activity relationship (QSAR) models for toxicity evaluation, it is difficult to validate the results generated from traditional QSAR models due to the lack of toxicity mechanism information of the prediction results. To improve the regulatory acceptance and model usefulness, our group developed mechanism-based models for the prediction of skin sensitization, developmental and reproductive toxicity, carcinogenicity, and neurotoxicity. While the models provide verifiable mechanism information with better performance and regulatory acceptance, only known mechanisms were considered individually in the models leading to potential false predictions. To provide complimentary toxicity information of a whole organism, we then developed a novel Virtual Zebrafish system to predict the overall influence on zebrafish including mortality, morphology, and behavioral effects. Zebrafish as a very important model organism can provide organism-level toxicity information to comprehend toxicity evaluation for complex endpoints. The Virtual Zebrafish system integrates QSAR and in silico toxicogenomics models to predict drug effects and ensure safety efficiently. The predicted effects can be further utilized to develop prediction methods and weight-of-evidence models for complex toxicity endpoints. The developed models for complex toxicity endpoints based on virtual zebrafish offer interpretable information of the mechanism that is favored by regulatory agencies. An online system will be developed to provide access to the developed models that can be applied to drug development and chemical hazard identification.