Seminars

Abstract: Long-read sequencing, aka Third Generation Sequencing/TGS (i.e., Oxford Nanopore Technologies/ONT and Pacific Biosciences/PacBio) can generate single-molecule long reads, ranging from a few kb to million bp. Long reads has been demonstrated to be very powerful to address many complex biomedical problems that remained unsolved by short reads. For example, the extensive applications of TGS data for genome research have been published in various biomedical contexts. Here, I will present the methodological research of how long reads can advance transcriptome research, and how their applications discovered new insights of transcript complexity and transposable elements in early embryonic development and stem cells. Especially, I will explain the theoretical basics of the quantitative analysis with long-read RNA-seq. When time allows, I will also talk about long-read epigenetics assay beyond merely methylation profiling.

Bio: Dr. Au is the professor of Computational Medicine and Bioinformatics at University of Michigan, Ann Arbor. He received a B.S. at Tsinghua University and Ph.D. of structural biology at University of Oxford. He also obtained a master’s degree of statistics and the postdoctoral training in Dr. Wing H. Wong’s group at Stanford University. In the early stage of his career, he published the first long-read RNA-seq methods plus bioinformatics software (PacBio platform: PNAS 2013; Nanopore platform: F1000research 2017). Dr. Au established his independent research team at University of Iowa in 2013 and extended his research to long read-based epigenetics research. Dr. Au relocated to The Ohio State University in 2018, where he served as the founding director of the PhD program of biomedical informatics and the Vice Chair for Research in the Department of Biomedical Informatics. Dr. Au’s research focuses on bioinformatics method development for sequencing data, especially for long reads. His team is also interested in developing innovative experimental long-read assays for (epi-)transcriptomics and epigenetics, and they are applying these experimental and bioinformatics techniques to interrogate the problems of transcript complexity and transposable elements in early embryonic development and stem cell biology.

Abstract: We are in the midst of an AI hypecycle. Large language models are everywhere you look – and they are increasingly being integrated into biomedical research and clinical care.  The potential upside for AI is sky high and yet, haven’t we been here before?  In this seminar, I will review why AI, and particularly machine learning, has become the technology du jour in biomedicine - again.  I will provide illustrations of machine learning in support of novel biomedical discovery, including drug repurposing and automation in clinical phenotyping.  At the same time, I will review how blind trust in AI can lead to numerous societal dilemmas, including violations of privacy, algorithmic unfairness, and an overall loss of trust.  I will then show how these problems be represented in AI development and application lifecyle, so that problems can be spotted and addressed before they destroy our research ecosystems and clinical operations.

Bio: Bradley Malin, Ph.D. is the Accenture Professor of Biomedical Informatics, Biostatistics, and Computer Science at Vanderbilt University Medical Center. He co-founded and co-directs ADVANCE Center, which is focused on the development of foundational AI models, their translation into biomedical research and clinical practice, and continuous monitoring and surveillance.  He is a principal investigator of the Instructure Core of the NIH AIM-AHEAD Program and the Ethical and Trustworthy AI Core of the NIH Bridge2AI Center.  He is a member of the Board of Scientific Counselors of the National Center for Health Statistics (NCHS) of the U.S. Centers for Disease Control and Prevention (CDC), as well as the Technical Anonymisation Group (TAG) of the European Medicines Agency.  Among various honors, he is an elected fellow of the National Academy of Medicine and was a recipient of the Presidential Early Career Award for Scientists and Engineers (PECASE)

Abstract: Rare variants significantly impact complex diseases. This presentation will first introduce SAIGE-GENE and SAIGE-GENE+, methodologies extending SAIGE to gene/region-based rare variant tests. These methods efficiently utilize mixed effects models to adjust for sample relatedness and saddlepoint approximations to account for case-control imbalance. SAIGE-GENE+ additionally incorporates functional annotations and collapsing of ultra-rare variants that can help to improve type I error control and power. In the second part of the talk, I will introduce our recent workto estimate effect sizes of rare variants. The method, RareEffect, uses an empirical Bayesian approach that estimates gene/region-level heritability and then an effect size of each variant. We also show the effect sizes obtained from our model can be leveraged to improve the performance of polygenic scores.  

Bio: Seunggeun (Shawn) Lee is a Professor of Data Science at Seoul National University. Before moving to his current position, he was an Associate Professor of Biostatistics at the University of Michigan. He received his PhD in Biostatistics from the University of North Carolina at Chapel and completed postdoctoral training at Harvard School of Public Health. His research focuses on developing statistical and computational methods for the analysis of large biobanks, which is essential to better understand the genetic architecture of complex diseases and traits.

Abstract: We are laying the foundations of AI for molecular discovery and drug design, eventually enabling AI to learn and innovate on its own. Instead of training separate models for every task, we leverage large language and geometric models across many tasks through fine-tuning and few-shot prompting. Central to our approach is the integration of molecular structures, biological knowledge, and genomic data into AI models. We are advancing self-supervised learning to leverage multi-omics datasets and geometric deep learning to model the geometry of molecules. I describe PINNACLE AI, contextual AI models for single-cell protein biology. PINNACLE models enhance 3D structures of protein-protein interactions in immune-oncology, predict drug effects across cell types and cell states, and nominate therapeutic targets in a cell-type-specific manner. For drugs to be effective, they must act on biological targets in core disease processes. I describe our multimodal sequence-structure generative models that design molecules to serve as optimal binders with desired biochemical properties. Finally, candidate drugs need to be matched to patient benefits. I introduce TxGNN, a knowledge graph AI model for zero-shot prediction of therapeutic use across over 17,000 diseases, enabling drug repurposing for 7,000 rare diseases, with a mere 5% having FDA-approved drugs. TxGNN's predictions align with medication use across millions of medical records.  

Bio: Marinka Zitnik (https://zitniklab.hms.harvard.edu) is an Assistant Professor of Biomedical Informatics at Harvard Medical School, at Kempner Institute for the Study of Natural and Artificial Intelligence at Harvard University and at Broad Institute of MIT and Harvard. Zitnik investigates foundations of AI that contribute to the scientific understanding of medicine and therapeutic design, eventually enabling AI to learn and innovate on its own. Her research won best paper and research awards, including the Kavli Fellowship of the National Academy of Sciences, Kaneb Fellowship award at Harvard Medical School, NSF CAREER Award, awards from the International Society for Computational Biology, International Conference in Machine Learning, Bayer Early Excellence in Science, Amazon Faculty Research, Google Faculty Research, Roche Alliance with Distinguished Scientists, and Sanofi iDEA-iTECH Award. Zitnik founded Therapeutics Data Commons, a global open-science initiative to access and evaluate AI across stages of development and therapeutic modalities, and she served as the faculty lead of the AI4Science initiative.