Sangwon Hyun
University of California-Santa Cruz, CA·Mathematics & Statistics
Phytoplankton, the microscopic drifting algae at the base of ocean food webs, vary from place to place in ways shaped by their environment. Flow cytometry measures the size and fluorescence of single cells as they stream past a laser.
- Fields
Mathematics & Statistics
Earth & Environmental Science
Computer Science
- Topics
flow cytometry
phytoplankton
marine microbes
- Grants
- 1 known grant, through Sep 2028
What they work on
- Develop computationally efficient mixture of neural network models that automatically classify flow cytometry cell data while accounting for environmental covariates, applied to over 100 Pacific and Atlantic cruises.
- Build a generative neural network framework for changepoint detection whose latent variables identify shifts in phytoplankton communities and help redefine ecological ocean provinces.
- Apply convolutional neural networks to density regression and spatial interpolation, predicting full cytograms beyond cruise tracks to build global phytoplankton maps.
More about the research
Research cruises collect huge amounts of these data, but analyzing them is slow and largely manual. This collaborative project builds machine learning and statistical models that turn data from more than 100 cruises across the Pacific and Atlantic Oceans into maps of where phytoplankton live and how the environment drives those patterns.
Why it matters
Automating the analysis of ocean flow cytometry data could let oceanographers map phytoplankton populations and traits across entire ocean basins, not just along ship tracks. This would show how marine microbes respond to changing environments. Public software could also help other fields that handle high-dimensional mixture data, such as environmental science, agriculture and epidemiology.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Jeff Bowman
University of California-San Diego, CA·Earth & Environmental Science
Identify the ecological conditions and microbial taxa that explain variation in microbial growth efficiency along light and nutrient gradients in the California Current ecosystem.
Aaron Maxwell
West Virginia University, WV·Earth & Environmental Science·Early-career
Develop convolutional neural network methods for semantic and instance segmentation to extract geomorphic features from LiDAR point clouds and terrain derivatives such as hillshades.
Erin McParland
Oregon State University, OR·Earth & Environmental Science
Quantify about 150 small polar extracellular metabolites in seawater across four regimes that differ in biological productivity and oxygen, using new metabolomics methods.
The school
University of California-Santa Cruz: 15 labs on PI Pages, $218M a year on research, 183 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.