Prof. Wing-Kin Sung

Chinese University of Hong Kong and Hong Kong Genome Institute

Professor Wing-Kin Sung is a Global Stem Professor in the Department of Chemical Pathology, the Chinese University of Hong Kong. He is the director of the Laboratory of Computational Genomics. He is also the Chief Bioinformatics officer (Honorary) in the Hong Kong Genome Institute. His recent research focuses on identifying genomic mutations from high-throughput sequencing data and on understanding the relationship between mutations (in particular, structural variations) and diseases. Prof. Sung received both the B.Sc. and the Ph.D. degree in the Department of Computer Science from the University of Hong Kong in 1993, 1998, respectively. He has over 25 years of experience in Algorithm and Bioinformatics research. Prior to joining CUHK, Professor Sung was a Professor in the Department of Computer Science at the National University of Singapore (NUS) and was a senior group leader at the Genome Institute of Singapore. He is an expert in the field of bioinformatics, who has been leading the development of a number of bioinformatics software and has over 290 high impact papers published in renowned academic journals, including Bioinformatics, Cell, Nature, Nature Genetics and Nucleic Acids Research. In recognition of his research contributions, Professor Sung was conferred the FIT Paper Award (Japan) in 2003, the National Science Award (Singapore) in 2006, and the Young Researcher Award (NUS) in 2008. He has also served in the programming committee for over 70 international conferences.

Speech Title: "Repeat-aware Structural Variation Calling and its Clinical Applications"

Abstract: Structural variations (SVs) are known to be crucial in disease development, and second-generation sequencing technologies have greatly aided their detection. However, existing software falls short, identifying only approximately 50% of SVs. A key observation is that the performance of current software is notably compromised when SVs are present in repeat regions. To address this limitation, we have developed some novel repeat-aware structural variation callers over the past decade. In this presentation, we will delve into the exploration of potential strategies to further enhance SV calling performance, particularly in these challenging scenarios involving repeat regions. Furthermore, we will highlight the practical application of our improved methods in the diagnosis of rare diseases.

 

Invited Speakers

 

Prof. Chanchal Mitra

University of Hyderabad

Chanchal Mitra did his Bachelors and Masters from the University of Calcutta and Ph.D. from the Tata Institute of Fundamental Research (University of Bombay). He did his post doctoral work at the State University of New York at Albany (The University at Albany), USA and also at the University of Lund, Sweden. His research interests are Bioinformatics, Computational Biology and Biosensors (enzyme based). He joined University of Hyderabad in 1985 as a lecturer and retired in 2015 as Professor of Biochemistry. He has supervised several Ph.D. students, project students and research associates. He has over 100 publications in peer reviewed journals. According to google scholar, he has citations 1272, h-index of 20 and i10-index of 31. He lives in Hyderabad, India (2023).

Speech Title: "Machine Learning in Bioinformatics: PCA and Disease"

Abstract: In this presentation, we shall discuss the foundations of Principal Component Analysis (PCA) and discuss with example how to predict the onset of a disease. We use the PIMA Indian database (https://www.kaggle.com) in this analysis. We have used the tools build in the mathematical and statistical package R (https://www.r-project.org) for principal component analysis. In this example we shall use prcomp function which is based on SVD (singular value decomposition) principle. We discuss how the dimensionality of the problem can be reduced from original eight to about four or five. We shall also discuss and explore the role of the size of the database (50 and 100 complete) on the noise or accuracy of the results. Finally we shall discuss the limitations of the method and the implicit assumptions made in the basic analysis. The focus will be on application and practical use.

 

Prof. Guan Ning Lin

Shanghai Jiao Tong University

Guan Ning Lin, a full Professor at Shanghai Jiao Tong University, China and a researcher PI at Shanghai Mental Health Center. He is the Deputy Director of the Engineering Research Center of Digital Medicine, Ministry of Education (ERCDM), and the Deputy Director of Imaging, Computing, and Systems Biology at the School of Biomedical Engineering, Shanghai Jiao Tong University. He was selected as the "Shanghai Oriental Scholar (Distinguished Professor)" by the Shanghai Municipal Education Commission in 2016, and currently has the National Natural Science Foundation Research Fund for Foreign Outstanding Young Scholars. His main focus is bioinformatics, systematic evaluation of blood biomarkers of mental diseases, and deep learning and development and application of clinical decision systems. He is a member of the Bioinformatics Special Committee of the China Computer Federation, a member of the Bone and Joint Special Committee of the Shanghai Biomedical Engineering Society, etc. His recent publications include Nature Human Behavior, Science Advances, Genome Biology, Neuron, Nature Communication, American Journal of Human Genetics, etc.

Speech Title: "Inferring the Effects of Protein Variants on Protein–protein Interactions with Interpretable Transformer Representations"

Abstract: Identifying pathogenetic variants and inferring their impact on protein–protein interactions sheds light on their functional consequences on diseases. Limited by the availability of experimental data on the consequences of protein interaction, most existing methods focus on building models to predict changes in protein binding affinity. In this talk, I will introduce MIPPI, an end-to-end, interpretable transformer-based deep learning model that learns features directly from sequences to determine the types of variant impact on protein-protein interactions. In addition, I will show the practicality of MIPPI in prioritizing de novo mutations from neurodevelopmental disorders and the potential to determine the pathogenic and driving mutations. Overall, MIPPI is a versatile, robust, and interpretable model, capable of effectively predicting mutation impacts on protein-protein interactions and facilitating the discovery of clinically actionable variants.

 

Assoc. Prof. Jie Zheng

ShanghaiTech University

Jie Zheng is an Associate Professor (tenured) at the School of Information Science and Technology, ShanghaiTech University, Shanghai, China. He received his B. Eng from Zhejiang University in China, and his Ph.D. from the University of California, Riverside in USA, both in Computer Science. He worked as a Postdoctoral Visiting Fellow and Research Associate at the National Center for Biotechnology Information (NCBI), National Library of Medicine (NLM), National Institutes of Health (NIH), USA, and as an Assistant Professor at the School of Computer Science and Engineering, Nanyang Technological University (NTU), Singapore. Dr. Zheng’s current research interests are bioinformatics and computational biology, biomedical data science, and AI for Science (focused on AI for drug discovery and AI for cancer precision medicine). He develops novel algorithmic and AI methods (e.g. machine learning and data science techniques, data-driven and knowledge-driven models) to help answer biomedical questions, such as how to precisely and selectively kill tumor cells.

Speech Title: "Explainable AI for Prediction of Synthetic Lethal Gene Pairs as Anti-cancer Drug Targets"

Abstract: Synthetic Lethality (SL) is a type of genetic interaction typically between two genes, which is that perturbations to both genes will kill a cell but perturbation to one gene will not. It is a gold mine of anti-cancer drug targets since targeting an SL partner of a gene with cancer-specific mutation can selectively kill cancer cells without harming normal cells. Current wet-lab screening methods usually have high cost, while statistical and machine learning methods cannot fully utilize the prior knowledge or lack clear explanations. We have developed a series of deep learning methods using Knowledge Graphs (KGs) and Explainable Artificial Intelligence (XAI) to predict SLs and understand mechanisms behind. SynLethDB is a comprehensive database including many SL gene pairs and a knowledge graph named SynLethKG. KG4SL is the first method using KG for SL prediction (ISMB/ECCB 2021). PiLSL further focuses on pairwise interaction learning from KGs for predicting SLs with interpretability (ECCB 2022). Recently, we proposed KR4SL, which employs path-based knowledge reasoning to rank SL candidate partners for given primary genes, and is able to explain the SL prediction and biological mechanisms clearly (ISMB/ECCB 2023). Our ongoing work combines pre-trained Language Models (e.g. GPT) and KGs to explain SLs in natural languages. In future, we aim to unlock the power of XAI by integrating more data and knowledge, and promoting closer collaborations with wet-lab biologists, clinical scientists and pharmaceutical researchers, to catalyze advancements of AI for cancer precision medicine.

 

Assoc. Prof. Hon-Cheong So

The Chinese University of Hong Kong, Hong Kong

Dr. Hon-Cheong So received his Bachelor of Medicine and Bachelor of Surgery (MBBS) degree together with a PhD degree in 2012 from The University of Hong Kong (HKU). His PhD research focused on statistical and psychiatric genomics. He has received numerous awards for his academic achievement, including the Croucher Foundation Scholarship and the Dr. Stephen K.P. Chang Gold Medal for the best PhD thesis in the Faculty of Medicine, HKU. Prior to taking up the current academic post, he worked as a resident psychiatrist in Queen Mary Hospital and Castle Peak Hospital. He joined the School of Biomedical Sciences of The Chinese University of Hong Kong as an Assistant Professor in Jan 2016 and is also currently assistant professor (by courtesy) of the Department of Psychiatry, CUHK. His main research interests include the development and application of novel statistical and computational methodologies to “omics” and clinical data in general. In particular, he is interested in uncovering the genetic architecture of complex diseases and predicting disease risk and phenotypes based on bioinformatics and clinical data.

Speech Title: "Machine Learning and Causal Inference Approaches to Understating Biomedical and Clinical Data"

Abstract: Recently years have witnessed exponential growth in biomedical and clinical ‘big data’. Machine learning (ML) approaches, unlike standard regression models, may capture complex non-linear and interactive relationships between different variables. We will highlight several of our works that have leveraged ML approaches, such as supervised ML methods for drug repositioning, prediction of drug targets, and predicting the severity of COVID-19 infection. We will also highlight applications of unsupervised learning such as disease subtyping. While statistical and computational methods may reveal associations easily, deciphering causal relationships between variables is more difficult. Causal inference has become an active research area in statistics and artificial intelligence. We will illustrate our recent research on the development and application of causal inference methods for clinical/biomedical research. Examples include the use of Mendelian randomization to reveal causal risk factors for complex diseases, and Bayesian network to uncover causal relationships between genes and clinical outcomes.

 

 

Previous Speakers

 

Prof. Yuan-Ting Zhang City University of Hong Kong	Prof. Alexander Suvoror Institute of Experimental Medicine, St. Petersburg	Prof. David Zhang The Chinese University of Hong Kong (Shenzhen)
Prof.Tun-Wen Pai National Taipei University of Technology	Prof. Dong-Qing Wei Shanghai Jiaotong University	Prof. TSUI Kwok-Wing Stephen The Chinese University of Hong Kong
Prof. Cathy Wu University of Delaware	Prof. Xuegong Zhang Tsinghua University	Prof. Yi Pan Chinese Academy of Sciences
Prof. Bairong Shen Sichuan University	Prof. Peiyu Zhang Henan University	Prof. Zheng Zhou Chinese Academy of Sciences
Prof. Le Zhang Sichuan University	Prof. Fei Guo Central South University	Prof. Bin Liu Beijing Institute of Technology
Mr. Xiaoqiang Li China National GeneBank DataBase