Serena Y. Yeung

CellFluxV2 is an image-generative model that predicts cell morphology changes under perturbations, achieving higher image fidelity than baselines and enabling batch-effect correction and cell-state interpolation.

The paper uses cryoEELS and cryoET to study calcium distribution in mitochondria of human retinal ganglion cells under calcifying conditions, revealing mitochondrial calcification.

CryoViT improves segmentation of large subcellular structures in cryoET data using vision transformers, outperforming U-Net-based methods with limited training data.

This study uses cryoET and proteomics to identify mitochondrial and organelle abnormalities in HD models, showing potential for early diagnosis and therapy development.

This paper introduces O2-VAE, an orientation-invariant representation learning method for morphologic profiling of cells and organelles, improving clustering and outlier detection.