Dr Yike Shen
Columbia University, USA
Abstract: High-dimensional data in environmental health sciences present unprecedented challenges and opportunities. In this presentation, I will share my research spanning from microbiome to using novel machine learning methods in toxicity and disease prediction. Humans are exposed to a myriad of environmental pollutants (e.g., pharmaceuticals and personal care products, metals, PFAS, flame retardants, etc.) every day. Individual or mixture of pollutants may relate to changes in the human gut microbiome and/or soil-plant microbiome, and subsequently impact human health. I will discuss our recent projects in the microbiome epidemiology including but not limited to the influences of environmental metals on children’s gut microbiome in a healthy gestation cohort. Altering the gut microbiome is not the only way environmental pollutants can impact human health. I will also talk about my doctoral work on antimicrobial resistance and bacterial microbiome in lettuce-soil systems in response to antibiotic exposure. Accurate prediction of fate and transport of contaminants in the environment is a critical step in chemical risk assessment, which often involves high-dimensional data. We recently developed an autoencoder deep learning models to learn latent space chemical representations of hundreds of physiochemical features to accurately predict chemical ecotoxicity. I will also talk about our successful prediction of hundreds of pesticide dissipation half-life intervals in plants using gradient boosted regression trees-molecular fingerprints. In the future, I will keep developing novel computational tools to untangle the complex relationships between high-dimensional data in exposure, omics, and health outcomes to protect human and environmental health.
Host：Assist. Prof. Zhenxing Mao
EEH Early Career Board Member
Zhengzhou University
Time：9:00pm Sept 15, 2022 (Beijing time)
Zoom ID: 816 9975 7155
Bilibili: 25002335
