Date: 15 December 2025 (Monday)
Time: 11:00 am – 12:00 pm
Venue: Lecture Theatre 2, G/F, William M.W. Mong Block, 21 Sassoon Road
Host: Professor Joshua Ho

Lana Garmire is a nationally and internationally known expert in biomedical informatics. She is a full professor and vice chair of research of the Department of Biomedical Informatics and Data Science, in the Medical School of University of Alabama Birmingham (UAB). She previously worked as faculty at U Michigan (2018-2025) and U of Hawaii (2012-2018). She obtained the MA degree in Statistics (2005) and PhD degree in Comparative Biochemistry (Computational Biology focus, 2007), both from UC-Berkeley. Over the years, her research expanded to multi-modal based survival prediction, single-cell and spatial bioinformatics, and Electronical Health Record based disease analysis and modeling. She has published over 100 papers, delivered over 100 invited talks to institutes including the National Library of Medicine (NLM) and National Academy of Sciences (NAS). She has mentored over 100 trainees in the research lab, ranging from assistant professors to undergraduates. She has served on various NIH study sections and was a standing member of BDMA study section. She is on the editorial advisory board for journals Genome Biology and GigaScience. Among the awards, notable ones include US Presidential Early Career Scientists and Engineers award in 2019. She is a fellow of American Institute of Medical and Biological Engineering (AIMBE) since 2022, and a fellow of American College of Medical Informatics (ACMI) since 2025.

Heterogeneity is an intrinsic property of multicellular organisms, and recognizing cell-to-cell variation is essential for translational biomedical research. In this talk, I will illustrate how biological heterogeneity can be rigorously addressed under two distinct objectives: DNA-methylome–based epigenetic biomarker discovery and drug repurposing. First, using preeclampsia as an example, I will demonstrate how adjusting for cell-type composition and relevant clinical covariates leads to a striking loss of initially identified CpG associations—highlighting the importance of correcting for underlying heterogeneity. Second, I will introduce ASGARD, a single-cell–based framework that leverages cell-type–specific information to predict repurposable drugs at the individual patient level. Together, these examples emphasize the analytical rigor required when working with heterogeneous biological data and encourage the audience to adopt more nuanced approaches in their own research. All are welcome.