Yang Joon Kim

This paper evaluates Tabular Foundation Models against specialized models for predicting cellular perturbation responses across multiple biological scales and tasks.

RNAquarium integrates zebrafish and metatranscriptomic data to identify viral drivers of transcriptomic states and support foundation-model training.

DanioDecima improves gene expression prediction in zebrafish by transferring knowledge from human and mouse models, enhancing correlations and generating synthetic promoters with expected lineage-specific motifs.

CZ CELLxGENE Discover provides a platform for scalable exploration and analysis of aggregated single-cell transcriptomic data across multiple studies.

The paper presents Zebrahub-Multiome, a single-cell multiomic atlas of zebrafish embryogenesis, revealing gene regulatory network dynamics and cis-regulatory elements across developmental stages.

The paper provides a framework and interactive software tools for manually annotating cell types in single-cell multi-omics datasets, demonstrating how different modalities can be combined to improve annotation accuracy.

Zebrafish embryos sequenced individually across ten developmental stages, paired with light-sheet lineage reconstruction, form an atlas and browser for in silico fate mapping.

The paper compares MERFISH spatial transcriptomics with bulk and single-cell RNA sequencing in mouse liver and kidney, finding MERFISH provides comparable results and robust cell type identification without needing integration with scRNA-seq atlases.