We are pleased to announce our upcoming webinar in November 2025. Dr. Kwangmoon Park from University of Pennsylvania will give a talk at 2pm (ET) on November 19 (Wednesday). Please use the link below to register for the KISS webinar. The webinar title and abstract are as follows.
Date/Time: 2pm – 3pm ET (1pm – 2pm CT; 11am – 12pm PT) on November 19
Registration link:
https://yonsei.zoom.us/meeting/register/JPXJDmI-RKOx6XT9tK0FFA
If the link does not open when clicked, please copy and paste it into your browser’s address bar.
Registration is required for this meeting. After registering, you will receive a confirmation email containing information about joining the meeting.
Speaker: Kwangmoon Park from University of Pennsylvania.
Title: Sensitivity Analysis and Outcome Dimension Reduction in Large-Scale, Observational Genomic Studies
Abstract: In large-scale observational genomic studies, such as single-cell RNA-sequencing experiments, multiple outcomes are of interest, and each outcome often exhibits different sensitivities to unmeasured confounding, where some outcomes are more sensitive to biases from an unmeasured confounder while others are less sensitive. It is also common to assume that the outcomes exhibit low-dimensional, low-rank structure. Leveraging both features of such datasets, we propose a novel procedure called LaunchODR, which conducts sensitivity-aware dimension reduction for testing treatment effects in observational studies with multivariate outcomes. Specifically, LaunchODR utilizes dimension reduction methods to identify the shared low-dimensional structure and conducts the “least sensitive test” for the average treatment effect across outcomes. Notably, the proposed procedure leverages a previously unexplored insight in observational studies: if the multivariate outcomes share a common low-dimensional structure, then the individual biases from unmeasured confounding also have (up to rotation) an identical low-dimensional structure. Under suitable assumptions, we show that LaunchODR asymptotically identifies the correct low-dimensional embedding, and the resulting test has Type I error control and is consistent. Finally, we demonstrate LaunchODR on a population-scale epigenomic dataset to investigate the causal effect of Amyloid-β (i.e., treatment) on multiple DNA methylation regions (i.e., outcomes). Our results show that the DNA methylation regions are significantly influenced by Amyloid-β levels even in the presence of potential unmeasured confounders, and they align with previous in vitro experiments.
I am looking forward to seeing you all!
Best regards,
Jeong Hoon Jang
KISS Program Chair Elect